[Engine-devel] Proposal VDSM <=> Engine Data Statistics Retrieval Optimization
This is a multi-part message in MIME format. --------------020305070807050907000005 Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Please find the prettier version on the wiki: http://www.ovirt.org/Proposal_VDSM_-_Engine_Data_Statistics_Retrieval Proposal VDSM - Engine Data Statistics Retrieval VDSM <=> Engine data retrieval optimization Motivation: Currently the RHEVM engine is polling the a lot of data from VDSM every 15 seconds. This should be optimized and the amount of data requested should be more specific. For each VM the data currently contains much more information than actually needed which blows up the size of the XML content quite big. We could optimize this by splitting the reply on the getVmStats based on the request of the engine into sections. For this reason Omer Frenkel and me have split up the data into parts based on their usage. This data can and usually does change during the lifetime of the VM. Rarely Changed: This data is change not very frequent and it should be enough to update this only once in a while. Most commonly this data changes after changes made in the UI or after a migration of the VM to another Host. *Status* = Running *acpiEnable* = true *vmType* = kvm *guestName* = W864GUESTAGENTT *displayType* = qxl *guestOs* = Win 8 *kvmEnable* = true #/*this should be constant and never changed*/ *pauseCode* = NOERR *monitorResponse* = 0 *session* = Locked # unused *netIfaces* = [{'name': 'Realtek RTL8139C+ Fast Ethernet NIC', 'inet6': ['fe80::490c:92bb:bbcc:9f87'], 'inet': ['10.34.60.148'], 'hw': '00:1a:4a:22:3c:db'}] *appsList* = ['RHEV-Tools 3.2.4', 'RHEV-Agent64 3.2.3', 'RHEV-Serial64 3.2.3', 'RHEV-Network64 3.2.2', 'RHEV-Network64 3.2.3', 'RHEV-Block64 3.2.3', 'RHEV-Balloon64 3.2.3', 'RHEV-Balloon64 3.2.2', 'RHEV-Agent64 3.2.2', 'RHEV-USB 3.2.3', 'RHEV-Block64 3.2.2', 'RHEV-Serial64 3.2.2'] *pid* = 11314 *guestIPs* = 10.34.60.148 # duplicated info *displayIp* = 0 *displayPort* = 5902 *displaySecurePort* = 5903 *username* = user@W864GUESTAGENTT *clientIp* = *lastLogin* = 1361976900.67 Often Changed: This data is changed quite often however it is not necessary to update this data every 15 seconds. As this is cumulative data and reflects the current status, and it does not need to be snapshotted every 15 seconds to retrieve statistics. The data can be retrieved in much more generous time slices. (e.g. Every 5 minutes) *network* = {'vnet1': {'macAddr': '00:1a:4a:22:3c:db', 'rxDropped': '0', 'txDropped': '0', 'rxErrors': '0', 'txRate': '0.0', 'rxRate': '0.0', 'txErrors': '0', 'state': 'unknown', 'speed': '100', 'name': 'vnet1'}} *disksUsage* = [{'path': 'c:\\', 'total': '64055406592', 'fs': 'NTFS', 'used': '19223846912'}, {'path': 'd:\\', 'total': '3490912256', 'fs': 'UDF', 'used': '3490912256'}] *timeOffset* = 14422 *elapsedTime* = 68591 *hash* = 2335461227228498964 *statsAge* = 0.09 # unused Often Changed but unused This data does not seem to be used in the engine at all. It is *not* even used in the data warehouse. *memoryStats* = {'swap_out': '0', 'majflt': '0', 'mem_free': '1466884', 'swap_in': '0', 'pageflt': '0', 'mem_total': '2096736', 'mem_unused': '1466884'} *balloonInfo* = {'balloon_max': 2097152, 'balloon_cur': 2097152} *disks* = {'vda': {'readLatency': '0', 'apparentsize': '64424509440', 'writeLatency': '1754496', 'imageID': '28abb923-7b89-4638-84f8-1700f0b76482', 'flushLatency': '156549', 'readRate': '0.00', 'truesize': '18855059456', 'writeRate': '952.05'}, 'hdc': {'readLatency': '0', 'apparentsize': '0', 'writeLatency': '0', 'flushLatency': '0', 'readRate': '0.00', 'truesize': '0', 'writeRate': '0.00'}} Very frequent uppdates needed by webadmin portal: This data is mostly needed for the webadmin portal and might be required to be updated quite often. An exception here is the statsAge field, which seems to be unused by the Engine. This data could be requested every 15 seconds to keep things as they are now. *cpuSys* = 2.32 *cpuUser* = 1.34 *memUsage* = 30 Proposed Solution for VDSM & Engine: We will introduce new optional parameters to getVmStats, getAllVmStats and list to allow a finer grained specification of data which should be included. *Parameter:* *statsType*=/*<string>*/ (getVmStats, getAllVmStats only) *Allowed values:* * full (default to keep backwards compatibility) * app-list (Just send the application list) * rare (include everything from rarely changed to very frequent) * often (include everything from often changed to very frequent) * frequent (only send the very frequently changed items) *Parameter:* *clientId*=*<string>* The client id is specified by the client and should be unique however constantly used. *Parameter:* *diff*=*<boolean>* In combination with the clientId VDSM will send only differences to the previous request from the named clientId. (if diff=true) Additional Change: Besides the introduction of the new parameters for list, getVmStats and getAllVmStats it might make sense to include a hash for the appList into the rarely changed section of the response which would allow to identify changes and avoid having to sent the complete appList every so often and only if the hash known to the client is outdated. *Note:* The appList (Application List) reported by the guest agent could be fully implemented on request only, as long as the guest agent installed supports this. As there seems to be a request to have the complete list of installed applications on all guests this data could be quite extensive and a huge list. On the other hand this data is only rarely visible and therefore it should not be requested all the time and only on demand. Improvement of the Guest Agent: As part of the proposed solution it is necessary to improve the guest agent as well. For the full application list there should be implemented a caching system which will be fully reactive and should not poll the application list for example all the time. The guest can create a prepared data file containing all data in the JSON format (as used for the communication with VDSM via VIO) and just have to read that file from disk and directly sends it to VDSM. However it is quite possible that this list is to big and it might have to be chunked into pieces. (Multiple messages, which would have to be supported by VDSM then as well) The solution for this is to make VDSM request this data and it will retrieve the data necessary on request only. -- Regards, Vinzenz Feenstra | Senior Software Engineer RedHat Engineering Virtualization R & D Phone: +420 532 294 625 IRC: vfeenstr or evilissimo Better technology. Faster innovation. Powered by community collaboration. See how it works at redhat.com --------------020305070807050907000005 Content-Type: text/html; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit <html> <head> <meta http-equiv="content-type" content="text/html; charset=ISO-8859-1"> </head> <body bgcolor="#FFFFFF" text="#000000"> Please find the prettier version on the wiki: <a class="moz-txt-link-freetext" href="http://www.ovirt.org/Proposal_VDSM_-_Engine_Data_Statistics_Retrieval">http://www.ovirt.org/Proposal_VDSM_-_Engine_Data_Statistics_Retrieval</a><br> <br> <h1 id="firstHeading" class="firstHeading page-header"><span dir="auto">Proposal VDSM - Engine Data Statistics Retrieval</span></h1> <h2><span class="mw-headline" id="VDSM_.3C.3D.3E_Engine_data_retrieval_optimization">VDSM <=> Engine data retrieval optimization </span></h2> <h3> <span class="mw-headline" id="Motivation:"> Motivation: </span></h3> <p>Currently the RHEVM engine is polling the a lot of data from VDSM every 15 seconds. This should be optimized and the amount of data requested should be more specific. </p> <p>For each VM the data currently contains much more information than actually needed which blows up the size of the XML content quite big. We could optimize this by splitting the reply on the getVmStats based on the request of the engine into sections. For this reason Omer Frenkel and me have split up the data into parts based on their usage. </p> <p>This data can and usually does change during the lifetime of the VM. </p> <h4> <span class="mw-headline" id="Rarely_Changed:"> Rarely Changed: </span></h4> <p>This data is change not very frequent and it should be enough to update this only once in a while. Most commonly this data changes after changes made in the UI or after a migration of the VM to another Host. </p> <pre> <b>Status</b> = Running <b>acpiEnable</b> = true <b>vmType</b> = kvm <b>guestName</b> = W864GUESTAGENTT <b>displayType</b> = qxl <b>guestOs</b> = Win 8 <b>kvmEnable</b> = true # <i><b>this should be constant and never changed</b></i> <b>pauseCode</b> = NOERR <b>monitorResponse</b> = 0 <b>session</b> = Locked # unused <b>netIfaces</b> = [{'name': 'Realtek RTL8139C+ Fast Ethernet NIC', 'inet6': ['fe80::490c:92bb:bbcc:9f87'], 'inet': ['10.34.60.148'], 'hw': '00:1a:4a:22:3c:db'}] <b>appsList</b> = ['RHEV-Tools 3.2.4', 'RHEV-Agent64 3.2.3', 'RHEV-Serial64 3.2.3', 'RHEV-Network64 3.2.2', 'RHEV-Network64 3.2.3', 'RHEV-Block64 3.2.3', 'RHEV-Balloon64 3.2.3', 'RHEV-Balloon64 3.2.2', 'RHEV-Agent64 3.2.2', 'RHEV-USB 3.2.3', 'RHEV-Block64 3.2.2', 'RHEV-Serial64 3.2.2'] <b>pid</b> = 11314 <b>guestIPs</b> = 10.34.60.148 # duplicated info <b>displayIp</b> = 0 <b>displayPort</b> = 5902 <b>displaySecurePort</b> = 5903 <b>username</b> = user@W864GUESTAGENTT <b>clientIp</b> = <b>lastLogin</b> = 1361976900.67 </pre> <h4> <span class="mw-headline" id="Often_Changed:"> Often Changed: </span></h4> <p>This data is changed quite often however it is not necessary to update this data every 15 seconds. As this is cumulative data and reflects the current status, and it does not need to be snapshotted every 15 seconds to retrieve statistics. The data can be retrieved in much more generous time slices. (e.g. Every 5 minutes) </p> <pre> <b>network</b> = {'vnet1': {'macAddr': '00:1a:4a:22:3c:db', 'rxDropped': '0', 'txDropped': '0', 'rxErrors': '0', 'txRate': '0.0', 'rxRate': '0.0', 'txErrors': '0', 'state': 'unknown', 'speed': '100', 'name': 'vnet1'}} <b>disksUsage</b> = [{'path': 'c:\\', 'total': '64055406592', 'fs': 'NTFS', 'used': '19223846912'}, {'path': 'd:\\', 'total': '3490912256', 'fs': 'UDF', 'used': '3490912256'}] <b>timeOffset</b> = 14422 <b>elapsedTime</b> = 68591 <b>hash</b> = 2335461227228498964 <b>statsAge</b> = 0.09 # unused </pre> <h4> <span class="mw-headline" id="Often_Changed_but_unused"> Often Changed but unused </span></h4> <p>This data does not seem to be used in the engine at all. It is <b>not</b> even used in the data warehouse. </p> <pre> <b>memoryStats</b> = {'swap_out': '0', 'majflt': '0', 'mem_free': '1466884', 'swap_in': '0', 'pageflt': '0', 'mem_total': '2096736', 'mem_unused': '1466884'} <b>balloonInfo</b> = {'balloon_max': 2097152, 'balloon_cur': 2097152} <b>disks</b> = {'vda': {'readLatency': '0', 'apparentsize': '64424509440', 'writeLatency': '1754496', 'imageID': '28abb923-7b89-4638-84f8-1700f0b76482', 'flushLatency': '156549', 'readRate': '0.00', 'truesize': '18855059456', 'writeRate': '952.05'}, 'hdc': {'readLatency': '0', 'apparentsize': '0', 'writeLatency': '0', 'flushLatency': '0', 'readRate': '0.00', 'truesize': '0', 'writeRate': '0.00'}} </pre> <h4> <span class="mw-headline" id="Very_frequent_uppdates_needed_by_webadmin_portal:"> Very frequent uppdates needed by webadmin portal: </span></h4> <p>This data is mostly needed for the webadmin portal and might be required to be updated quite often. An exception here is the statsAge field, which seems to be unused by the Engine. This data could be requested every 15 seconds to keep things as they are now. </p> <pre> <b>cpuSys</b> = 2.32 <b>cpuUser</b> = 1.34 <b>memUsage</b> = 30 </pre> <h2> <span class="mw-headline" id="Proposed_Solution_for_VDSM_.26_Engine:"> Proposed Solution for VDSM & Engine: </span></h2> <p>We will introduce new optional parameters to getVmStats, getAllVmStats and list to allow a finer grained specification of data which should be included. </p> <p><b>Parameter:</b> <b>statsType</b>=<i><b><string></b></i> (getVmStats, getAllVmStats only) <b>Allowed values:</b> </p> <ul> <li> full (default to keep backwards compatibility) </li> <li> app-list (Just send the application list) </li> <li> rare (include everything from rarely changed to very frequent) </li> <li> often (include everything from often changed to very frequent) </li> <li> frequent (only send the very frequently changed items) </li> </ul> <p><br> <b>Parameter:</b> <b>clientId</b>=<b><string></b> The client id is specified by the client and should be unique however constantly used. </p> <p><b>Parameter:</b> <b>diff</b>=<b><boolean></b> In combination with the clientId VDSM will send only differences to the previous request from the named clientId. (if diff=true) </p> <p><br> </p> <h3> <span class="mw-headline" id="Additional_Change:"> Additional Change: </span></h3> <p>Besides the introduction of the new parameters for list, getVmStats and getAllVmStats it might make sense to include a hash for the appList into the rarely changed section of the response which would allow to identify changes and avoid having to sent the complete appList every so often and only if the hash known to the client is outdated. </p> <p><b>Note:</b> The appList (Application List) reported by the guest agent could be fully implemented on request only, as long as the guest agent installed supports this. As there seems to be a request to have the complete list of installed applications on all guests this data could be quite extensive and a huge list. On the other hand this data is only rarely visible and therefore it should not be requested all the time and only on demand. </p> <h3> <span class="mw-headline" id="Improvement_of_the_Guest_Agent:"> Improvement of the Guest Agent: </span></h3> <p>As part of the proposed solution it is necessary to improve the guest agent as well. For the full application list there should be implemented a caching system which will be fully reactive and should not poll the application list for example all the time. The guest can create a prepared data file containing all data in the JSON format (as used for the communication with VDSM via VIO) and just have to read that file from disk and directly sends it to VDSM. However it is quite possible that this list is to big and it might have to be chunked into pieces. (Multiple messages, which would have to be supported by VDSM then as well) The solution for this is to make VDSM request this data and it will retrieve the data necessary on request only. </p> <pre class="moz-signature" cols="72">-- Regards, Vinzenz Feenstra | Senior Software Engineer RedHat Engineering Virtualization R & D Phone: +420 532 294 625 IRC: vfeenstr or evilissimo Better technology. Faster innovation. Powered by community collaboration. See how it works at redhat.com</pre> </body> </html> --------------020305070807050907000005--
On Thu, Mar 07, 2013 at 12:25:54PM +0100, Vinzenz Feenstra wrote:
Please find the prettier version on the wiki: http://www.ovirt.org/Proposal_VDSM_-_Engine_Data_Statistics_Retrieval
Proposal VDSM - Engine Data Statistics Retrieval
VDSM <=> Engine data retrieval optimization
Motivation:
Currently the RHEVM engine is polling the a lot of data from VDSM every 15 seconds. This should be optimized and the amount of data requested should be more specific.
It feels like a good idea, but do you have numbers? How much traffic would be saved? Remember the added computation incurred on each host - there's always a price to pay.
For each VM the data currently contains much more information than actually needed which blows up the size of the XML content quite big. We could optimize this by splitting the reply on the getVmStats based on the request of the engine into sections. For this reason Omer Frenkel and me have split up the data into parts based on their usage.
This data can and usually does change during the lifetime of the VM.
Rarely Changed:
This data is change not very frequent and it should be enough to update this only once in a while. Most commonly this data changes after changes made in the UI or after a migration of the VM to another Host.
*Status* = Running
Status does not change much, but when it does, it is important to report that quickly.
*acpiEnable* = true *vmType* = kvm *guestName* = W864GUESTAGENTT *displayType* = qxl *guestOs* = Win 8 *kvmEnable* = true #/*this should be constant and never changed*/ *pauseCode* = NOERR *monitorResponse* = 0 *session* = Locked # unused *netIfaces* = [{'name': 'Realtek RTL8139C+ Fast Ethernet NIC', 'inet6': ['fe80::490c:92bb:bbcc:9f87'], 'inet': ['10.34.60.148'], 'hw': '00:1a:4a:22:3c:db'}] *appsList* = ['RHEV-Tools 3.2.4', 'RHEV-Agent64 3.2.3', 'RHEV-Serial64 3.2.3', 'RHEV-Network64 3.2.2', 'RHEV-Network64 3.2.3', 'RHEV-Block64 3.2.3', 'RHEV-Balloon64 3.2.3', 'RHEV-Balloon64 3.2.2', 'RHEV-Agent64 3.2.2', 'RHEV-USB 3.2.3', 'RHEV-Block64 3.2.2', 'RHEV-Serial64 3.2.2'] *pid* = 11314 *guestIPs* = 10.34.60.148 # duplicated info *displayIp* = 0 *displayPort* = 5902 *displaySecurePort* = 5903 *username* = user@W864GUESTAGENTT *clientIp* = *lastLogin* = 1361976900.67
Often Changed:
This data is changed quite often however it is not necessary to update this data every 15 seconds. As this is cumulative data and reflects the current status, and it does not need to be snapshotted every 15 seconds to retrieve statistics. The data can be retrieved in much more generous time slices. (e.g. Every 5 minutes)
*network* = {'vnet1': {'macAddr': '00:1a:4a:22:3c:db', 'rxDropped': '0', 'txDropped': '0', 'rxErrors': '0', 'txRate': '0.0', 'rxRate': '0.0', 'txErrors': '0', 'state': 'unknown', 'speed': '100', 'name': 'vnet1'}} *disksUsage* = [{'path': 'c:\\', 'total': '64055406592', 'fs': 'NTFS', 'used': '19223846912'}, {'path': 'd:\\', 'total': '3490912256', 'fs': 'UDF', 'used': '3490912256'}] *timeOffset* = 14422 *elapsedTime* = 68591 *hash* = 2335461227228498964 *statsAge* = 0.09 # unused
Often Changed but unused
This data does not seem to be used in the engine at all. It is *not* even used in the data warehouse.
*memoryStats* = {'swap_out': '0', 'majflt': '0', 'mem_free': '1466884', 'swap_in': '0', 'pageflt': '0', 'mem_total': '2096736', 'mem_unused': '1466884'} *balloonInfo* = {'balloon_max': 2097152, 'balloon_cur': 2097152} *disks* = {'vda': {'readLatency': '0', 'apparentsize': '64424509440', 'writeLatency': '1754496', 'imageID': '28abb923-7b89-4638-84f8-1700f0b76482', 'flushLatency': '156549', 'readRate': '0.00', 'truesize': '18855059456', 'writeRate': '952.05'}, 'hdc': {'readLatency': '0', 'apparentsize': '0', 'writeLatency': '0', 'flushLatency': '0', 'readRate': '0.00', 'truesize': '0', 'writeRate': '0.00'}}
I am pretty sure that {read,write,flush}Latency is collected and reported by Engine. `git grep writeLatency` reinforces my vague memory.
Very frequent uppdates needed by webadmin portal:
This data is mostly needed for the webadmin portal and might be required to be updated quite often. An exception here is the statsAge field, which seems to be unused by the Engine. This data could be requested every 15 seconds to keep things as they are now.
*cpuSys* = 2.32 *cpuUser* = 1.34 *memUsage* = 30
Proposed Solution for VDSM & Engine:
We will introduce new optional parameters to getVmStats, getAllVmStats and list to allow a finer grained specification of data which should be included.
*Parameter:* *statsType*=/*<string>*/ (getVmStats, getAllVmStats only) *Allowed values:*
* full (default to keep backwards compatibility) * app-list (Just send the application list) * rare (include everything from rarely changed to very frequent) * often (include everything from often changed to very frequent) * frequent (only send the very frequently changed items)
I think that a nice way to think of this, is that Engine ask for a set of keys it is interested about. Asking for getVmStats(keys=[displayType, netIfaces]) would return only the requrested values of the VM. "full", "rare", "often" and "frequent" are simply pre-defined sets of key names. A side effect of this pov is that we can avoid the vague name "statsType".
*Parameter:* *clientId*=*<string>* The client id is specified by the client and should be unique however constantly used.
*Parameter:* *diff*=*<boolean>* In combination with the clientId VDSM will send only differences to the previous request from the named clientId. (if diff=true)
The semantics of "diff" is not completely defined: how about complex structures like that of "network"? It is most likely to be reported every time. Since this requires a caching mechanism on vdsm side, Engine must expect that the cache may be evicted in any moment, and that a full list is received.
Additional Change:
Besides the introduction of the new parameters for list, getVmStats and getAllVmStats it might make sense to include a hash for the appList into the rarely changed section of the response which would allow to identify changes and avoid having to sent the complete appList every so often and only if the hash known to the client is outdated.
*Note:* The appList (Application List) reported by the guest agent could be fully implemented on request only, as long as the guest agent installed supports this. As there seems to be a request to have the complete list of installed applications on all guests this data could be quite extensive and a huge list. On the other hand this data is only rarely visible and therefore it should not be requested all the time and only on demand.
Improvement of the Guest Agent:
As part of the proposed solution it is necessary to improve the guest agent as well.
Improving the agent may be a good idea, but I do not see the necessity in it. It's also important to improve the horrible multithreaded vdsm/libvirt statistics acquisition, but just as unrelated to the core of this feature.
For the full application list there should be implemented a caching system which will be fully reactive and should not poll the application list for example all the time. The guest can create a prepared data file containing all data in the JSON format (as used for the communication with VDSM via VIO) and just have to read that file from disk and directly sends it to VDSM. However it is quite possible that this list is to big and it might have to be chunked into pieces. (Multiple messages, which would have to be supported by VDSM then as well) The solution for this is to make VDSM request this data and it will retrieve the data necessary on request only.
On 03/08/2013 06:11 AM, Dan Kenigsberg wrote:
Please find the prettier version on the wiki: http://www.ovirt.org/Proposal_VDSM_-_Engine_Data_Statistics_Retrieval
Proposal VDSM - Engine Data Statistics Retrieval
VDSM <=> Engine data retrieval optimization
Motivation:
Currently the RHEVM engine is polling the a lot of data from VDSM every 15 seconds. This should be optimized and the amount of data requested should be more specific. It feels like a good idea, but do you have numbers? How much traffic would be saved? Remember the added computation incurred on each host -
On Thu, Mar 07, 2013 at 12:25:54PM +0100, Vinzenz Feenstra wrote: there's always a price to pay.
For each VM the data currently contains much more information than actually needed which blows up the size of the XML content quite big. We could optimize this by splitting the reply on the getVmStats based on the request of the engine into sections. For this reason Omer Frenkel and me have split up the data into parts based on their usage.
This data can and usually does change during the lifetime of the VM.
Rarely Changed:
This data is change not very frequent and it should be enough to update this only once in a while. Most commonly this data changes after changes made in the UI or after a migration of the VM to another Host.
*Status* = Running Status does not change much, but when it does, it is important to report that quickly. For this kind of data, it is suitable to use an event report, which should be available in the jsonrpc API.
*acpiEnable* = true *vmType* = kvm *guestName* = W864GUESTAGENTT *displayType* = qxl *guestOs* = Win 8 *kvmEnable* = true #/*this should be constant and never changed*/ *pauseCode* = NOERR *monitorResponse* = 0 *session* = Locked # unused *netIfaces* = [{'name': 'Realtek RTL8139C+ Fast Ethernet NIC', 'inet6': ['fe80::490c:92bb:bbcc:9f87'], 'inet': ['10.34.60.148'], 'hw': '00:1a:4a:22:3c:db'}] *appsList* = ['RHEV-Tools 3.2.4', 'RHEV-Agent64 3.2.3', 'RHEV-Serial64 3.2.3', 'RHEV-Network64 3.2.2', 'RHEV-Network64 3.2.3', 'RHEV-Block64 3.2.3', 'RHEV-Balloon64 3.2.3', 'RHEV-Balloon64 3.2.2', 'RHEV-Agent64 3.2.2', 'RHEV-USB 3.2.3', 'RHEV-Block64 3.2.2', 'RHEV-Serial64 3.2.2'] *pid* = 11314 *guestIPs* = 10.34.60.148 # duplicated info *displayIp* = 0 *displayPort* = 5902 *displaySecurePort* = 5903 *username* = user@W864GUESTAGENTT *clientIp* = *lastLogin* = 1361976900.67
Often Changed:
This data is changed quite often however it is not necessary to update this data every 15 seconds. As this is cumulative data and reflects the current status, and it does not need to be snapshotted every 15 seconds to retrieve statistics. The data can be retrieved in much more generous time slices. (e.g. Every 5 minutes)
*network* = {'vnet1': {'macAddr': '00:1a:4a:22:3c:db', 'rxDropped': '0', 'txDropped': '0', 'rxErrors': '0', 'txRate': '0.0', 'rxRate': '0.0', 'txErrors': '0', 'state': 'unknown', 'speed': '100', 'name': 'vnet1'}} *disksUsage* = [{'path': 'c:\\', 'total': '64055406592', 'fs': 'NTFS', 'used': '19223846912'}, {'path': 'd:\\', 'total': '3490912256', 'fs': 'UDF', 'used': '3490912256'}] *timeOffset* = 14422 *elapsedTime* = 68591 *hash* = 2335461227228498964 *statsAge* = 0.09 # unused
Often Changed but unused
This data does not seem to be used in the engine at all. It is *not* even used in the data warehouse.
*memoryStats* = {'swap_out': '0', 'majflt': '0', 'mem_free': '1466884', 'swap_in': '0', 'pageflt': '0', 'mem_total': '2096736', 'mem_unused': '1466884'} *balloonInfo* = {'balloon_max': 2097152, 'balloon_cur': 2097152} *disks* = {'vda': {'readLatency': '0', 'apparentsize': '64424509440', 'writeLatency': '1754496', 'imageID': '28abb923-7b89-4638-84f8-1700f0b76482', 'flushLatency': '156549', 'readRate': '0.00', 'truesize': '18855059456', 'writeRate': '952.05'}, 'hdc': {'readLatency': '0', 'apparentsize': '0', 'writeLatency': '0', 'flushLatency': '0', 'readRate': '0.00', 'truesize': '0', 'writeRate': '0.00'}}
I am pretty sure that {read,write,flush}Latency is collected and reported by Engine. `git grep writeLatency` reinforces my vague memory.
Very frequent uppdates needed by webadmin portal:
This data is mostly needed for the webadmin portal and might be required to be updated quite often. An exception here is the statsAge field, which seems to be unused by the Engine. This data could be requested every 15 seconds to keep things as they are now.
*cpuSys* = 2.32 *cpuUser* = 1.34 *memUsage* = 30
Proposed Solution for VDSM & Engine:
We will introduce new optional parameters to getVmStats, getAllVmStats and list to allow a finer grained specification of data which should be included.
*Parameter:* *statsType*=/*<string>*/ (getVmStats, getAllVmStats only) *Allowed values:*
* full (default to keep backwards compatibility) * app-list (Just send the application list) * rare (include everything from rarely changed to very frequent) * often (include everything from often changed to very frequent) * frequent (only send the very frequently changed items)
I think that a nice way to think of this, is that Engine ask for a set of keys it is interested about. Asking for getVmStats(keys=[displayType, netIfaces]) would return only the requrested values of the VM. +1. It could split the information according to different functions, not just change frequency. "full", "rare", "often" and "frequent" are simply pre-defined sets of key names.
A side effect of this pov is that we can avoid the vague name "statsType".
*Parameter:* *clientId*=*<string>* The client id is specified by the client and should be unique however constantly used.
*Parameter:* *diff*=*<boolean>* In combination with the clientId VDSM will send only differences to the previous request from the named clientId. (if diff=true)
The semantics of "diff" is not completely defined: how about complex structures like that of "network"? It is most likely to be reported every time.
Since this requires a caching mechanism on vdsm side, Engine must expect that the cache may be evicted in any moment, and that a full list is received. Every data collector should be responsible to invalidate/update the cache. It could reduce the time to calculate the diff.
Additional Change:
Besides the introduction of the new parameters for list, getVmStats and getAllVmStats it might make sense to include a hash for the appList into the rarely changed section of the response which would allow to identify changes and avoid having to sent the complete appList every so often and only if the hash known to the client is outdated.
*Note:* The appList (Application List) reported by the guest agent could be fully implemented on request only, as long as the guest agent installed supports this. As there seems to be a request to have the complete list of installed applications on all guests this data could be quite extensive and a huge list. On the other hand this data is only rarely visible and therefore it should not be requested all the time and only on demand.
Improvement of the Guest Agent:
As part of the proposed solution it is necessary to improve the guest agent as well.
Improving the agent may be a good idea, but I do not see the necessity in it. It's also important to improve the horrible multithreaded vdsm/libvirt statistics acquisition, but just as unrelated to the core of this feature.
For the full application list there should be implemented a caching system which will be fully reactive and should not poll the application list for example all the time. The guest can create a prepared data file containing all data in the JSON format (as used for the communication with VDSM via VIO) and just have to read that file from disk and directly sends it to VDSM. However it is quite possible that this list is to big and it might have to be chunked into pieces. (Multiple messages, which would have to be supported by VDSM then as well) The solution for this is to make VDSM request this data and it will retrieve the data necessary on request only.
vdsm-devel mailing list vdsm-devel@lists.fedorahosted.org https://lists.fedorahosted.org/mailman/listinfo/vdsm-devel
On 03/08/2013 06:11 AM, Dan Kenigsberg wrote:
Please find the prettier version on the wiki: http://www.ovirt.org/Proposal_VDSM_-_Engine_Data_Statistics_Retrieval
Proposal VDSM - Engine Data Statistics Retrieval
VDSM <=> Engine data retrieval optimization
Motivation:
Currently the RHEVM engine is polling the a lot of data from VDSM every 15 seconds. This should be optimized and the amount of data requested should be more specific. It feels like a good idea, but do you have numbers? How much traffic would be saved? Remember the added computation incurred on each host -
On Thu, Mar 07, 2013 at 12:25:54PM +0100, Vinzenz Feenstra wrote: there's always a price to pay. Well the data of a single really basic simple VM has about 4 KiB data in
On 03/08/2013 03:30 AM, Mark Wu wrote: the output of vdsClient, the XMLRPC formatted body part has almost 16KiB. The thing is that this data is queried every 15 seconds (previously 10) with little value for having ALL data sent all the time, the engine is not even using all of the data all the time. This optimization must be seen on a bigger scale, if you have a datacenter with let's say 1000 VMs then the data needed to be transmitted and parsed by the engine every 15 seconds is about 16MiB. This optimization wouldn't pay off that much in a 2 server 20 VM datacenter however on a larger scale it has quite a big impact.
For each VM the data currently contains much more information than actually needed which blows up the size of the XML content quite big. We could optimize this by splitting the reply on the getVmStats based on the request of the engine into sections. For this reason Omer Frenkel and me have split up the data into parts based on their usage.
This data can and usually does change during the lifetime of the VM.
Rarely Changed:
This data is change not very frequent and it should be enough to update this only once in a while. Most commonly this data changes after changes made in the UI or after a migration of the VM to another Host.
*Status* = Running Status does not change much, but when it does, it is important to report that quickly.
This is done by the list command which is executed every 2 seconds (maybe 3?) For this kind of data, it is suitable to use an event report, which should be available in the jsonrpc API.
*acpiEnable* = true *vmType* = kvm *guestName* = W864GUESTAGENTT *displayType* = qxl *guestOs* = Win 8 *kvmEnable* = true #/*this should be constant and never changed*/ *pauseCode* = NOERR *monitorResponse* = 0 *session* = Locked # unused *netIfaces* = [{'name': 'Realtek RTL8139C+ Fast Ethernet NIC', 'inet6': ['fe80::490c:92bb:bbcc:9f87'], 'inet': ['10.34.60.148'], 'hw': '00:1a:4a:22:3c:db'}] *appsList* = ['RHEV-Tools 3.2.4', 'RHEV-Agent64 3.2.3', 'RHEV-Serial64 3.2.3', 'RHEV-Network64 3.2.2', 'RHEV-Network64 3.2.3', 'RHEV-Block64 3.2.3', 'RHEV-Balloon64 3.2.3', 'RHEV-Balloon64 3.2.2', 'RHEV-Agent64 3.2.2', 'RHEV-USB 3.2.3', 'RHEV-Block64 3.2.2', 'RHEV-Serial64 3.2.2'] *pid* = 11314 *guestIPs* = 10.34.60.148 # duplicated info *displayIp* = 0 *displayPort* = 5902 *displaySecurePort* = 5903 *username* = user@W864GUESTAGENTT *clientIp* = *lastLogin* = 1361976900.67
Often Changed:
This data is changed quite often however it is not necessary to update this data every 15 seconds. As this is cumulative data and reflects the current status, and it does not need to be snapshotted every 15 seconds to retrieve statistics. The data can be retrieved in much more generous time slices. (e.g. Every 5 minutes)
*network* = {'vnet1': {'macAddr': '00:1a:4a:22:3c:db', 'rxDropped': '0', 'txDropped': '0', 'rxErrors': '0', 'txRate': '0.0', 'rxRate': '0.0', 'txErrors': '0', 'state': 'unknown', 'speed': '100', 'name': 'vnet1'}} *disksUsage* = [{'path': 'c:\\', 'total': '64055406592', 'fs': 'NTFS', 'used': '19223846912'}, {'path': 'd:\\', 'total': '3490912256', 'fs': 'UDF', 'used': '3490912256'}] *timeOffset* = 14422 *elapsedTime* = 68591 *hash* = 2335461227228498964 *statsAge* = 0.09 # unused
Often Changed but unused
This data does not seem to be used in the engine at all. It is *not* even used in the data warehouse.
*memoryStats* = {'swap_out': '0', 'majflt': '0', 'mem_free': '1466884', 'swap_in': '0', 'pageflt': '0', 'mem_total': '2096736', 'mem_unused': '1466884'} *balloonInfo* = {'balloon_max': 2097152, 'balloon_cur': 2097152} *disks* = {'vda': {'readLatency': '0', 'apparentsize': '64424509440', 'writeLatency': '1754496', 'imageID': '28abb923-7b89-4638-84f8-1700f0b76482', 'flushLatency': '156549', 'readRate': '0.00', 'truesize': '18855059456', 'writeRate': '952.05'}, 'hdc': {'readLatency': '0', 'apparentsize': '0', 'writeLatency': '0', 'flushLatency': '0', 'readRate': '0.00', 'truesize': '0', 'writeRate': '0.00'}}
I am pretty sure that {read,write,flush}Latency is collected and reported by Engine. `git grep writeLatency` reinforces my vague memory.
Ok, well we did just a quite quick query about the usage and we searched rather for the keys than for the individual entries. Good to know what we need to be a bit more specific about the individual entries to classify them more appropriate.
Very frequent uppdates needed by webadmin portal:
This data is mostly needed for the webadmin portal and might be required to be updated quite often. An exception here is the statsAge field, which seems to be unused by the Engine. This data could be requested every 15 seconds to keep things as they are now.
*cpuSys* = 2.32 *cpuUser* = 1.34 *memUsage* = 30
Proposed Solution for VDSM & Engine:
We will introduce new optional parameters to getVmStats, getAllVmStats and list to allow a finer grained specification of data which should be included.
*Parameter:* *statsType*=/*<string>*/ (getVmStats, getAllVmStats only) *Allowed values:*
* full (default to keep backwards compatibility) * app-list (Just send the application list) * rare (include everything from rarely changed to very frequent) * often (include everything from often changed to very frequent) * frequent (only send the very frequently changed items)
I think that a nice way to think of this, is that Engine ask for a set of keys it is interested about. Asking for getVmStats(keys=[displayType, netIfaces]) would return only the requrested values of the VM. I was thinking of that as well or a way to exclude things from the list. +1. It could split the information according to different functions, not just change frequency. I would say to go for either or, both wouldn't make much sense. "full", "rare", "often" and "frequent" are simply pre-defined sets of key names.
A side effect of this pov is that we can avoid the vague name "statsType".
*Parameter:* *clientId*=*<string>* The client id is specified by the client and should be unique however constantly used.
*Parameter:* *diff*=*<boolean>* In combination with the clientId VDSM will send only differences to the previous request from the named clientId. (if diff=true)
The semantics of "diff" is not completely defined: how about complex structures like that of "network"? It is most likely to be reported every time. Well the idea was a per key evaluation, maybe in cases like network and disks per device/interface.
Since this requires a caching mechanism on vdsm side, Engine must expect that the cache may be evicted in any moment, and that a full list is received. Well the engine should always expect that. Every data collector should be responsible to invalidate/update the cache. It could reduce the time to calculate the diff.
Additional Change:
Besides the introduction of the new parameters for list, getVmStats and getAllVmStats it might make sense to include a hash for the appList into the rarely changed section of the response which would allow to identify changes and avoid having to sent the complete appList every so often and only if the hash known to the client is outdated.
*Note:* The appList (Application List) reported by the guest agent could be fully implemented on request only, as long as the guest agent installed supports this. As there seems to be a request to have the complete list of installed applications on all guests this data could be quite extensive and a huge list. On the other hand this data is only rarely visible and therefore it should not be requested all the time and only on demand.
Improvement of the Guest Agent:
As part of the proposed solution it is necessary to improve the guest agent as well.
Improving the agent may be a good idea, but I do not see the necessity in it. The guest agent is doing 'expensive' queries (e.g. "application_list") way too often. And things like network interfaces, disk usage and installed applications won't usually change every n minutes. Those queries could be much more reactive then proactive. It's also important to improve the horrible multithreaded vdsm/libvirt statistics acquisition, but just as unrelated to the core of this feature.
For the full application list there should be implemented a caching system which will be fully reactive and should not poll the application list for example all the time. The guest can create a prepared data file containing all data in the JSON format (as used for the communication with VDSM via VIO) and just have to read that file from disk and directly sends it to VDSM. However it is quite possible that this list is to big and it might have to be chunked into pieces. (Multiple messages, which would have to be supported by VDSM then as well) The solution for this is to make VDSM request this data and it will retrieve the data necessary on request only.
vdsm-devel mailing list vdsm-devel@lists.fedorahosted.org https://lists.fedorahosted.org/mailman/listinfo/vdsm-devel
-- Regards, Vinzenz Feenstra | Senior Software Engineer RedHat Engineering Virtualization R & D Phone: +420 532 294 625 IRC: vfeenstr or evilissimo Better technology. Faster innovation. Powered by community collaboration. See how it works at redhat.com
On Fri, Mar 08, 2013 at 09:37:03AM +0100, Vinzenz Feenstra wrote: <snip>
Improvement of the Guest Agent:
As part of the proposed solution it is necessary to improve the guest agent as well.
Improving the agent may be a good idea, but I do not see the necessity in it. The guest agent is doing 'expensive' queries (e.g. "application_list") way too often. And things like network interfaces, disk usage and installed applications won't usually change every n minutes. Those queries could be much more reactive then proactive.
Of course, but as I said:
It's also important to improve the horrible multithreaded vdsm/libvirt statistics acquisition, but just as unrelated to the core of this feature.
I think it is misleading to include this in the discussion about vdsm/Engine interface change.
On 03/08/2013 12:11 AM, Dan Kenigsberg wrote:
update this only once in a while. Most commonly this data changes after changes made in the UI or after a migration of the VM to another Host.
*Status* = Running Status does not change much, but when it does, it is important to report
This data is change not very frequent and it should be enough to that quickly.
actually, status and one other field are polled every 3 seconds iirc regardless of getAllVmStats (list table or something like that)
This is a multi-part message in MIME format. --------------050703010307000605090101 Content-Type: text/plain; charset=UTF-8; format=flowed Content-Transfer-Encoding: 7bit On 03/07/2013 07:25 PM, Vinzenz Feenstra wrote:
Please find the prettier version on the wiki: http://www.ovirt.org/Proposal_VDSM_-_Engine_Data_Statistics_Retrieval
Proposal VDSM - Engine Data Statistics Retrieval
VDSM <=> Engine data retrieval optimization
Motivation:
Currently the RHEVM engine is polling the a lot of data from VDSM every 15 seconds. This should be optimized and the amount of data requested should be more specific.
If the data size really matters, we could also consider to pack the information into binary. I am not sure if it's suitable in the transmission of XMLRPC.
For each VM the data currently contains much more information than actually needed which blows up the size of the XML content quite big. We could optimize this by splitting the reply on the getVmStats based on the request of the engine into sections. For this reason Omer Frenkel and me have split up the data into parts based on their usage.
This data can and usually does change during the lifetime of the VM.
Rarely Changed:
This data is change not very frequent and it should be enough to update this only once in a while. Most commonly this data changes after changes made in the UI or after a migration of the VM to another Host.
*Status* = Running *acpiEnable* = true *vmType* = kvm *guestName* = W864GUESTAGENTT *displayType* = qxl *guestOs* = Win 8 *kvmEnable* = true #/*this should be constant and never changed*/
Then it should be removed from vm stats. In my opinion, any information belongs to vm's static configuration, it shouldn't be included in vm stats. For the fields above, except 'Status', engine can get the information without querying the vdsm host. It could not be changed by vdsm itself, right?
*pauseCode* = NOERR *monitorResponse* = 0 *session* = Locked # unused *netIfaces* = [{'name': 'Realtek RTL8139C+ Fast Ethernet NIC', 'inet6': ['fe80::490c:92bb:bbcc:9f87'], 'inet': ['10.34.60.148'], 'hw': '00:1a:4a:22:3c:db'}] *appsList* = ['RHEV-Tools 3.2.4', 'RHEV-Agent64 3.2.3', 'RHEV-Serial64 3.2.3', 'RHEV-Network64 3.2.2', 'RHEV-Network64 3.2.3', 'RHEV-Block64 3.2.3', 'RHEV-Balloon64 3.2.3', 'RHEV-Balloon64 3.2.2', 'RHEV-Agent64 3.2.2', 'RHEV-USB 3.2.3', 'RHEV-Block64 3.2.2', 'RHEV-Serial64 3.2.2'] *pid* = 11314 *guestIPs* = 10.34.60.148 # duplicated info
*displayIp* = 0 *displayPort* = 5902 *displaySecurePort* = 5903
*username* = user@W864GUESTAGENTT *clientIp* = *lastLogin* = 1361976900.67
Often Changed:
This data is changed quite often however it is not necessary to update this data every 15 seconds. As this is cumulative data and reflects the current status, and it does not need to be snapshotted every 15 seconds to retrieve statistics. The data can be retrieved in much more generous time slices. (e.g. Every 5 minutes)
*network* = {'vnet1': {'macAddr': '00:1a:4a:22:3c:db', 'rxDropped': '0', 'txDropped': '0', 'rxErrors': '0', 'txRate': '0.0', 'rxRate': '0.0', 'txErrors': '0', 'state': 'unknown', 'speed': '100', 'name': 'vnet1'}}
macAddr and name don't change either.
*disksUsage* = [{'path': 'c:\\', 'total': '64055406592', 'fs': 'NTFS', 'used': '19223846912'}, {'path': 'd:\\', 'total': '3490912256', 'fs': 'UDF', 'used': '3490912256'}]
*timeOffset* = 14422 *elapsedTime* = 68591 *hash* = 2335461227228498964 *statsAge* = 0.09 # unused
Often Changed but unused
This data does not seem to be used in the engine at all. It is *not* even used in the data warehouse.
*memoryStats* = {'swap_out': '0', 'majflt': '0', 'mem_free': '1466884', 'swap_in': '0', 'pageflt': '0', 'mem_total': '2096736', 'mem_unused': '1466884'} *balloonInfo* = {'balloon_max': 2097152, 'balloon_cur': 2097152}
It's used by mom to adjust memory overcommitment dynamically.
*disks* = {'vda': {'readLatency': '0', 'apparentsize': '64424509440', 'writeLatency': '1754496', 'imageID': '28abb923-7b89-4638-84f8-1700f0b76482', 'flushLatency': '156549', 'readRate': '0.00', 'truesize': '18855059456', 'writeRate': '952.05'}, 'hdc': {'readLatency': '0', 'apparentsize': '0', 'writeLatency': '0', 'flushLatency': '0', 'readRate': '0.00', 'truesize': '0', 'writeRate': '0.00'}}
Very frequent uppdates needed by webadmin portal:
This data is mostly needed for the webadmin portal and might be required to be updated quite often. An exception here is the statsAge field, which seems to be unused by the Engine. This data could be requested every 15 seconds to keep things as they are now.
*cpuSys* = 2.32 *cpuUser* = 1.34 *memUsage* = 30
Proposed Solution for VDSM & Engine:
We will introduce new optional parameters to getVmStats, getAllVmStats and list to allow a finer grained specification of data which should be included.
*Parameter:* *statsType*=/*<string>*/ (getVmStats, getAllVmStats only) *Allowed values:*
* full (default to keep backwards compatibility) * app-list (Just send the application list) * rare (include everything from rarely changed to very frequent) * often (include everything from often changed to very frequent) * frequent (only send the very frequently changed items)
*Parameter:* *clientId*=*<string>* The client id is specified by the client and should be unique however constantly used.
*Parameter:* *diff*=*<boolean>* In combination with the clientId VDSM will send only differences to the previous request from the named clientId. (if diff=true)
Additional Change:
Besides the introduction of the new parameters for list, getVmStats and getAllVmStats it might make sense to include a hash for the appList into the rarely changed section of the response which would allow to identify changes and avoid having to sent the complete appList every so often and only if the hash known to the client is outdated.
*Note:* The appList (Application List) reported by the guest agent could be fully implemented on request only, as long as the guest agent installed supports this. As there seems to be a request to have the complete list of installed applications on all guests this data could be quite extensive and a huge list. On the other hand this data is only rarely visible and therefore it should not be requested all the time and only on demand.
Improvement of the Guest Agent:
As part of the proposed solution it is necessary to improve the guest agent as well. For the full application list there should be implemented a caching system which will be fully reactive and should not poll the application list for example all the time. The guest can create a prepared data file containing all data in the JSON format (as used for the communication with VDSM via VIO) and just have to read that file from disk and directly sends it to VDSM. However it is quite possible that this list is to big and it might have to be chunked into pieces. (Multiple messages, which would have to be supported by VDSM then as well) The solution for this is to make VDSM request this data and it will retrieve the data necessary on request only.
-- Regards,
Vinzenz Feenstra | Senior Software Engineer RedHat Engineering Virtualization R & D Phone: +420 532 294 625 IRC: vfeenstr or evilissimo
Better technology. Faster innovation. Powered by community collaboration. See how it works at redhat.com
_______________________________________________ vdsm-devel mailing list vdsm-devel@lists.fedorahosted.org https://lists.fedorahosted.org/mailman/listinfo/vdsm-devel
--------------050703010307000605090101 Content-Type: text/html; charset=UTF-8 Content-Transfer-Encoding: 8bit <html> <head> <meta content="text/html; charset=UTF-8" http-equiv="Content-Type"> </head> <body bgcolor="#FFFFFF" text="#000000"> <div class="moz-cite-prefix">On 03/07/2013 07:25 PM, Vinzenz Feenstra wrote:<br> </div> <blockquote cite="mid:51387942.606@redhat.com" type="cite"> <meta http-equiv="content-type" content="text/html; charset=UTF-8"> Please find the prettier version on the wiki: <a moz-do-not-send="true" class="moz-txt-link-freetext" href="http://www.ovirt.org/Proposal_VDSM_-_Engine_Data_Statistics_Retrieval">http://www.ovirt.org/Proposal_VDSM_-_Engine_Data_Statistics_Retrieval</a><br> <br> <h1 id="firstHeading" class="firstHeading page-header"><span dir="auto">Proposal VDSM - Engine Data Statistics Retrieval</span></h1> <h2><span class="mw-headline" id="VDSM_.3C.3D.3E_Engine_data_retrieval_optimization">VDSM <=> Engine data retrieval optimization </span></h2> <h3> <span class="mw-headline" id="Motivation:"> Motivation: </span></h3> <p>Currently the RHEVM engine is polling the a lot of data from VDSM every 15 seconds. This should be optimized and the amount of data requested should be more specific. </p> </blockquote> If the data size really matters, we could also consider to pack the information into binary. I am not sure if it's suitable in the <meta http-equiv="content-type" content="text/html; charset=UTF-8"> transmission of XMLRPC.<br> <blockquote cite="mid:51387942.606@redhat.com" type="cite"> <p>For each VM the data currently contains much more information than actually needed which blows up the size of the XML content quite big. We could optimize this by splitting the reply on the getVmStats based on the request of the engine into sections. For this reason Omer Frenkel and me have split up the data into parts based on their usage. </p> <p>This data can and usually does change during the lifetime of the VM. </p> <h4> <span class="mw-headline" id="Rarely_Changed:"> Rarely Changed: </span></h4> <p>This data is change not very frequent and it should be enough to update this only once in a while. Most commonly this data changes after changes made in the UI or after a migration of the VM to another Host. </p> <pre> <b>Status</b> = Running <b>acpiEnable</b> = true <b>vmType</b> = kvm <b>guestName</b> = W864GUESTAGENTT <b>displayType</b> = qxl <b>guestOs</b> = Win 8 <b>kvmEnable</b> = true # <i><b>this should be constant and never changed</b></i></pre> </blockquote> Then it should be removed from vm stats. In my opinion, any information belongs to vm's static configuration, it shouldn't be included in vm stats. For the fields above, except 'Status', engine can get the information without querying the vdsm host. It could not be changed by vdsm itself, right?<br> <blockquote cite="mid:51387942.606@redhat.com" type="cite"> <pre> <b>pauseCode</b> = NOERR <b>monitorResponse</b> = 0 <b>session</b> = Locked # unused <b>netIfaces</b> = [{'name': 'Realtek RTL8139C+ Fast Ethernet NIC', 'inet6': ['fe80::490c:92bb:bbcc:9f87'], 'inet': ['10.34.60.148'], 'hw': '00:1a:4a:22:3c:db'}] <b>appsList</b> = ['RHEV-Tools 3.2.4', 'RHEV-Agent64 3.2.3', 'RHEV-Serial64 3.2.3', 'RHEV-Network64 3.2.2', 'RHEV-Network64 3.2.3', 'RHEV-Block64 3.2.3', 'RHEV-Balloon64 3.2.3', 'RHEV-Balloon64 3.2.2', 'RHEV-Agent64 3.2.2', 'RHEV-USB 3.2.3', 'RHEV-Block64 3.2.2', 'RHEV-Serial64 3.2.2'] <b>pid</b> = 11314 <b>guestIPs</b> = 10.34.60.148 # duplicated info <b>displayIp</b> = 0 <b>displayPort</b> = 5902 <b>displaySecurePort</b> = 5903 <b>username</b> = user@W864GUESTAGENTT <b>clientIp</b> = <b>lastLogin</b> = 1361976900.67 </pre> <h4> <span class="mw-headline" id="Often_Changed:"> Often Changed: </span></h4> <p>This data is changed quite often however it is not necessary to update this data every 15 seconds. As this is cumulative data and reflects the current status, and it does not need to be snapshotted every 15 seconds to retrieve statistics. The data can be retrieved in much more generous time slices. (e.g. Every 5 minutes) </p> <pre> <b>network</b> = {'vnet1': {'macAddr': '00:1a:4a:22:3c:db', 'rxDropped': '0', 'txDropped': '0', 'rxErrors': '0', 'txRate': '0.0', 'rxRate': '0.0', 'txErrors': '0', 'state': 'unknown', 'speed': '100', 'name': 'vnet1'}}</pre> </blockquote> macAddr and name don't change either.<br> <blockquote cite="mid:51387942.606@redhat.com" type="cite"> <pre> <b>disksUsage</b> = [{'path': 'c:\\', 'total': '64055406592', 'fs': 'NTFS', 'used': '19223846912'}, {'path': 'd:\\', 'total': '3490912256', 'fs': 'UDF', 'used': '3490912256'}] <b>timeOffset</b> = 14422 <b>elapsedTime</b> = 68591 <b>hash</b> = 2335461227228498964 <b>statsAge</b> = 0.09 # unused </pre> <h4> <span class="mw-headline" id="Often_Changed_but_unused"> Often Changed but unused </span></h4> <p>This data does not seem to be used in the engine at all. It is <b>not</b> even used in the data warehouse. </p> <pre> <b>memoryStats</b> = {'swap_out': '0', 'majflt': '0', 'mem_free': '1466884', 'swap_in': '0', 'pageflt': '0', 'mem_total': '2096736', 'mem_unused': '1466884'} <b>balloonInfo</b> = {'balloon_max': 2097152, 'balloon_cur': 2097152} </pre> </blockquote> It's used by mom to adjust memory overcommitment dynamically.<br> <blockquote cite="mid:51387942.606@redhat.com" type="cite"> <pre> <b>disks</b> = {'vda': {'readLatency': '0', 'apparentsize': '64424509440', 'writeLatency': '1754496', 'imageID': '28abb923-7b89-4638-84f8-1700f0b76482', 'flushLatency': '156549', 'readRate': '0.00', 'truesize': '18855059456', 'writeRate': '952.05'}, 'hdc': {'readLatency': '0', 'apparentsize': '0', 'writeLatency': '0', 'flushLatency': '0', 'readRate': '0.00', 'truesize': '0', 'writeRate': '0.00'}} </pre> <h4> <span class="mw-headline" id="Very_frequent_uppdates_needed_by_webadmin_portal:"> Very frequent uppdates needed by webadmin portal: </span></h4> <p>This data is mostly needed for the webadmin portal and might be required to be updated quite often. An exception here is the statsAge field, which seems to be unused by the Engine. This data could be requested every 15 seconds to keep things as they are now. </p> <pre> <b>cpuSys</b> = 2.32 <b>cpuUser</b> = 1.34 <b>memUsage</b> = 30 </pre> <h2> <span class="mw-headline" id="Proposed_Solution_for_VDSM_.26_Engine:"> Proposed Solution for VDSM & Engine: </span></h2> <p>We will introduce new optional parameters to getVmStats, getAllVmStats and list to allow a finer grained specification of data which should be included. </p> <p><b>Parameter:</b> <b>statsType</b>=<i><b><string></b></i> (getVmStats, getAllVmStats only) <b>Allowed values:</b> </p> <ul> <li> full (default to keep backwards compatibility) </li> <li> app-list (Just send the application list) </li> <li> rare (include everything from rarely changed to very frequent) </li> <li> often (include everything from often changed to very frequent) </li> <li> frequent (only send the very frequently changed items) </li> </ul> <p><br> <b>Parameter:</b> <b>clientId</b>=<b><string></b> The client id is specified by the client and should be unique however constantly used. </p> <p><b>Parameter:</b> <b>diff</b>=<b><boolean></b> In combination with the clientId VDSM will send only differences to the previous request from the named clientId. (if diff=true) </p> <p><br> </p> <h3> <span class="mw-headline" id="Additional_Change:"> Additional Change: </span></h3> <p>Besides the introduction of the new parameters for list, getVmStats and getAllVmStats it might make sense to include a hash for the appList into the rarely changed section of the response which would allow to identify changes and avoid having to sent the complete appList every so often and only if the hash known to the client is outdated. </p> <p><b>Note:</b> The appList (Application List) reported by the guest agent could be fully implemented on request only, as long as the guest agent installed supports this. As there seems to be a request to have the complete list of installed applications on all guests this data could be quite extensive and a huge list. On the other hand this data is only rarely visible and therefore it should not be requested all the time and only on demand. </p> <h3> <span class="mw-headline" id="Improvement_of_the_Guest_Agent:"> Improvement of the Guest Agent: </span></h3> <p>As part of the proposed solution it is necessary to improve the guest agent as well. For the full application list there should be implemented a caching system which will be fully reactive and should not poll the application list for example all the time. The guest can create a prepared data file containing all data in the JSON format (as used for the communication with VDSM via VIO) and just have to read that file from disk and directly sends it to VDSM. However it is quite possible that this list is to big and it might have to be chunked into pieces. (Multiple messages, which would have to be supported by VDSM then as well) The solution for this is to make VDSM request this data and it will retrieve the data necessary on request only. </p> <pre class="moz-signature" cols="72">-- Regards, Vinzenz Feenstra | Senior Software Engineer RedHat Engineering Virtualization R & D Phone: +420 532 294 625 IRC: vfeenstr or evilissimo Better technology. Faster innovation. Powered by community collaboration. See how it works at redhat.com</pre> <br> <fieldset class="mimeAttachmentHeader"></fieldset> <br> <pre wrap="">_______________________________________________ vdsm-devel mailing list <a class="moz-txt-link-abbreviated" href="mailto:vdsm-devel@lists.fedorahosted.org">vdsm-devel@lists.fedorahosted.org</a> <a class="moz-txt-link-freetext" href="https://lists.fedorahosted.org/mailman/listinfo/vdsm-devel">https://lists.fedorahosted.org/mailman/listinfo/vdsm-devel</a> </pre> </blockquote> <br> </body> </html> --------------050703010307000605090101--
On Fri, Mar 08, 2013 at 10:21:27AM +0800, Mark Wu wrote:
On 03/07/2013 07:25 PM, Vinzenz Feenstra wrote:
Please find the prettier version on the wiki: http://www.ovirt.org/Proposal_VDSM_-_Engine_Data_Statistics_Retrieval
Proposal VDSM - Engine Data Statistics Retrieval
VDSM <=> Engine data retrieval optimization
Motivation:
Currently the RHEVM engine is polling the a lot of data from VDSM every 15 seconds. This should be optimized and the amount of data requested should be more specific.
If the data size really matters, we could also consider to pack the information into binary. I am not sure if it's suitable in the transmission of XMLRPC.
I do not think we should embed binary in XMLRPC. I'd consider compressing the data at the transport layer - but that would be a completely deferent feature.
For each VM the data currently contains much more information than actually needed which blows up the size of the XML content quite big. We could optimize this by splitting the reply on the getVmStats based on the request of the engine into sections. For this reason Omer Frenkel and me have split up the data into parts based on their usage.
This data can and usually does change during the lifetime of the VM.
Rarely Changed:
This data is change not very frequent and it should be enough to update this only once in a while. Most commonly this data changes after changes made in the UI or after a migration of the VM to another Host.
*Status* = Running *acpiEnable* = true *vmType* = kvm *guestName* = W864GUESTAGENTT *displayType* = qxl *guestOs* = Win 8 *kvmEnable* = true #/*this should be constant and never changed*/
Then it should be removed from vm stats. In my opinion, any information belongs to vm's static configuration, it shouldn't be included in vm stats. For the fields above, except 'Status', engine can get the information without querying the vdsm host. It could not be changed by vdsm itself, right?
actually, guestName and guestOs may change - for example by installing Linux on that Windows guest.
I am completely against this. It make the return value differ according to input which is a big no no when talking about type safe APIs. The only reason we have this problem is because there is this thing against making multiple calls. Just split it up. getVmRuntimeStats() - transient things like mem and cpu% getVmInformation() - (semi)static things like disk\networking layout etc. Each updated at different intervals. ----- Original Message -----
From: "Vinzenz Feenstra" <vfeenstr@redhat.com> To: vdsm-devel@lists.fedorahosted.org, engine-devel@ovirt.org Sent: Thursday, March 7, 2013 6:25:54 AM Subject: [Engine-devel] Proposal VDSM <=> Engine Data Statistics Retrieval Optimization
Please find the prettier version on the wiki: http://www.ovirt.org/Proposal_VDSM_-_Engine_Data_Statistics_Retrieval
Proposal VDSM - Engine Data Statistics Retrieval VDSM <=> Engine data retrieval optimization Motivation:
Currently the RHEVM engine is polling the a lot of data from VDSM every 15 seconds. This should be optimized and the amount of data requested should be more specific.
For each VM the data currently contains much more information than actually needed which blows up the size of the XML content quite big. We could optimize this by splitting the reply on the getVmStats based on the request of the engine into sections. For this reason Omer Frenkel and me have split up the data into parts based on their usage.
This data can and usually does change during the lifetime of the VM. Rarely Changed:
This data is change not very frequent and it should be enough to update this only once in a while. Most commonly this data changes after changes made in the UI or after a migration of the VM to another Host. Status = Running acpiEnable = true vmType = kvm guestName = W864GUESTAGENTT displayType = qxl guestOs = Win 8 kvmEnable = true # this should be constant and never changed pauseCode = NOERR monitorResponse = 0 session = Locked # unused netIfaces = [{'name': 'Realtek RTL8139C+ Fast Ethernet NIC', 'inet6': ['fe80::490c:92bb:bbcc:9f87'], 'inet': ['10.34.60.148'], 'hw': '00:1a:4a:22:3c:db'}] appsList = ['RHEV-Tools 3.2.4', 'RHEV-Agent64 3.2.3', 'RHEV-Serial64 3.2.3', 'RHEV-Network64 3.2.2', 'RHEV-Network64 3.2.3', 'RHEV-Block64 3.2.3', 'RHEV-Balloon64 3.2.3', 'RHEV-Balloon64 3.2.2', 'RHEV-Agent64 3.2.2', 'RHEV-USB 3.2.3', 'RHEV-Block64 3.2.2', 'RHEV-Serial64 3.2.2'] pid = 11314 guestIPs = 10.34.60.148 # duplicated info displayIp = 0 displayPort = 5902 displaySecurePort = 5903 username = user@W864GUESTAGENTT clientIp = lastLogin = 1361976900.67 Often Changed:
This data is changed quite often however it is not necessary to update this data every 15 seconds. As this is cumulative data and reflects the current status, and it does not need to be snapshotted every 15 seconds to retrieve statistics. The data can be retrieved in much more generous time slices. (e.g. Every 5 minutes) network = {'vnet1': {'macAddr': '00:1a:4a:22:3c:db', 'rxDropped': '0', 'txDropped': '0', 'rxErrors': '0', 'txRate': '0.0', 'rxRate': '0.0', 'txErrors': '0', 'state': 'unknown', 'speed': '100', 'name': 'vnet1'}} disksUsage = [{'path': 'c:\\', 'total': '64055406592', 'fs': 'NTFS', 'used': '19223846912'}, {'path': 'd:\\', 'total': '3490912256', 'fs': 'UDF', 'used': '3490912256'}] timeOffset = 14422 elapsedTime = 68591 hash = 2335461227228498964 statsAge = 0.09 # unused Often Changed but unused
This data does not seem to be used in the engine at all. It is not even used in the data warehouse. memoryStats = {'swap_out': '0', 'majflt': '0', 'mem_free': '1466884', 'swap_in': '0', 'pageflt': '0', 'mem_total': '2096736', 'mem_unused': '1466884'} balloonInfo = {'balloon_max': 2097152, 'balloon_cur': 2097152} disks = {'vda': {'readLatency': '0', 'apparentsize': '64424509440', 'writeLatency': '1754496', 'imageID': '28abb923-7b89-4638-84f8-1700f0b76482', 'flushLatency': '156549', 'readRate': '0.00', 'truesize': '18855059456', 'writeRate': '952.05'}, 'hdc': {'readLatency': '0', 'apparentsize': '0', 'writeLatency': '0', 'flushLatency': '0', 'readRate': '0.00', 'truesize': '0', 'writeRate': '0.00'}} Very frequent uppdates needed by webadmin portal:
This data is mostly needed for the webadmin portal and might be required to be updated quite often. An exception here is the statsAge field, which seems to be unused by the Engine. This data could be requested every 15 seconds to keep things as they are now. cpuSys = 2.32 cpuUser = 1.34 memUsage = 30 Proposed Solution for VDSM & Engine:
We will introduce new optional parameters to getVmStats, getAllVmStats and list to allow a finer grained specification of data which should be included.
Parameter: statsType = <string> (getVmStats, getAllVmStats only) Allowed values:
* full (default to keep backwards compatibility) * app-list (Just send the application list) * rare (include everything from rarely changed to very frequent) * often (include everything from often changed to very frequent) * frequent (only send the very frequently changed items)
Parameter: clientId = <string> The client id is specified by the client and should be unique however constantly used.
Parameter: diff = <boolean> In combination with the clientId VDSM will send only differences to the previous request from the named clientId. (if diff=true)
Additional Change:
Besides the introduction of the new parameters for list, getVmStats and getAllVmStats it might make sense to include a hash for the appList into the rarely changed section of the response which would allow to identify changes and avoid having to sent the complete appList every so often and only if the hash known to the client is outdated.
Note: The appList (Application List) reported by the guest agent could be fully implemented on request only, as long as the guest agent installed supports this. As there seems to be a request to have the complete list of installed applications on all guests this data could be quite extensive and a huge list. On the other hand this data is only rarely visible and therefore it should not be requested all the time and only on demand. Improvement of the Guest Agent:
As part of the proposed solution it is necessary to improve the guest agent as well. For the full application list there should be implemented a caching system which will be fully reactive and should not poll the application list for example all the time. The guest can create a prepared data file containing all data in the JSON format (as used for the communication with VDSM via VIO) and just have to read that file from disk and directly sends it to VDSM. However it is quite possible that this list is to big and it might have to be chunked into pieces. (Multiple messages, which would have to be supported by VDSM then as well) The solution for this is to make VDSM request this data and it will retrieve the data necessary on request only. -- Regards,
Vinzenz Feenstra | Senior Software Engineer RedHat Engineering Virtualization R & D Phone: +420 532 294 625 IRC: vfeenstr or evilissimo
Better technology. Faster innovation. Powered by community collaboration. See how it works at redhat.com _______________________________________________ Engine-devel mailing list Engine-devel@ovirt.org http://lists.ovirt.org/mailman/listinfo/engine-devel
----- Original Message -----
I am completely against this. It make the return value differ according to input which is a big no no when talking about type safe APIs.
The only reason we have this problem is because there is this thing against making multiple calls.
Just split it up. getVmRuntimeStats() - transient things like mem and cpu% getVmInformation() - (semi)static things like disk\networking layout etc. Each updated at different intervals.
+1 on splitting the data up into 2 separate API calls. You could potentially add a checksum (md5, or any other way) of the "static" data to getVmRuntimeStats and not bother even with polling the VmInformation if this hasn't changed. Then you could poll as often as you'd like the stats and immediately see if you also need to retrieve VmInfo or not (you rarely would).
----- Original Message -----
From: "Vinzenz Feenstra" <vfeenstr@redhat.com> To: vdsm-devel@lists.fedorahosted.org, engine-devel@ovirt.org Sent: Thursday, March 7, 2013 6:25:54 AM Subject: [Engine-devel] Proposal VDSM <=> Engine Data Statistics Retrieval Optimization
Please find the prettier version on the wiki: http://www.ovirt.org/Proposal_VDSM_-_Engine_Data_Statistics_Retrieval
Proposal VDSM - Engine Data Statistics Retrieval VDSM <=> Engine data retrieval optimization Motivation:
Currently the RHEVM engine is polling the a lot of data from VDSM every 15 seconds. This should be optimized and the amount of data requested should be more specific.
For each VM the data currently contains much more information than actually needed which blows up the size of the XML content quite big. We could optimize this by splitting the reply on the getVmStats based on the request of the engine into sections. For this reason Omer Frenkel and me have split up the data into parts based on their usage.
This data can and usually does change during the lifetime of the VM. Rarely Changed:
This data is change not very frequent and it should be enough to update this only once in a while. Most commonly this data changes after changes made in the UI or after a migration of the VM to another Host. Status = Running acpiEnable = true vmType = kvm guestName = W864GUESTAGENTT displayType = qxl guestOs = Win 8 kvmEnable = true # this should be constant and never changed pauseCode = NOERR monitorResponse = 0 session = Locked # unused netIfaces = [{'name': 'Realtek RTL8139C+ Fast Ethernet NIC', 'inet6': ['fe80::490c:92bb:bbcc:9f87'], 'inet': ['10.34.60.148'], 'hw': '00:1a:4a:22:3c:db'}] appsList = ['RHEV-Tools 3.2.4', 'RHEV-Agent64 3.2.3', 'RHEV-Serial64 3.2.3', 'RHEV-Network64 3.2.2', 'RHEV-Network64 3.2.3', 'RHEV-Block64 3.2.3', 'RHEV-Balloon64 3.2.3', 'RHEV-Balloon64 3.2.2', 'RHEV-Agent64 3.2.2', 'RHEV-USB 3.2.3', 'RHEV-Block64 3.2.2', 'RHEV-Serial64 3.2.2'] pid = 11314 guestIPs = 10.34.60.148 # duplicated info displayIp = 0 displayPort = 5902 displaySecurePort = 5903 username = user@W864GUESTAGENTT clientIp = lastLogin = 1361976900.67 Often Changed:
This data is changed quite often however it is not necessary to update this data every 15 seconds. As this is cumulative data and reflects the current status, and it does not need to be snapshotted every 15 seconds to retrieve statistics. The data can be retrieved in much more generous time slices. (e.g. Every 5 minutes) network = {'vnet1': {'macAddr': '00:1a:4a:22:3c:db', 'rxDropped': '0', 'txDropped': '0', 'rxErrors': '0', 'txRate': '0.0', 'rxRate': '0.0', 'txErrors': '0', 'state': 'unknown', 'speed': '100', 'name': 'vnet1'}} disksUsage = [{'path': 'c:\\', 'total': '64055406592', 'fs': 'NTFS', 'used': '19223846912'}, {'path': 'd:\\', 'total': '3490912256', 'fs': 'UDF', 'used': '3490912256'}] timeOffset = 14422 elapsedTime = 68591 hash = 2335461227228498964 statsAge = 0.09 # unused Often Changed but unused
This data does not seem to be used in the engine at all. It is not even used in the data warehouse. memoryStats = {'swap_out': '0', 'majflt': '0', 'mem_free': '1466884', 'swap_in': '0', 'pageflt': '0', 'mem_total': '2096736', 'mem_unused': '1466884'} balloonInfo = {'balloon_max': 2097152, 'balloon_cur': 2097152} disks = {'vda': {'readLatency': '0', 'apparentsize': '64424509440', 'writeLatency': '1754496', 'imageID': '28abb923-7b89-4638-84f8-1700f0b76482', 'flushLatency': '156549', 'readRate': '0.00', 'truesize': '18855059456', 'writeRate': '952.05'}, 'hdc': {'readLatency': '0', 'apparentsize': '0', 'writeLatency': '0', 'flushLatency': '0', 'readRate': '0.00', 'truesize': '0', 'writeRate': '0.00'}} Very frequent uppdates needed by webadmin portal:
This data is mostly needed for the webadmin portal and might be required to be updated quite often. An exception here is the statsAge field, which seems to be unused by the Engine. This data could be requested every 15 seconds to keep things as they are now. cpuSys = 2.32 cpuUser = 1.34 memUsage = 30 Proposed Solution for VDSM & Engine:
We will introduce new optional parameters to getVmStats, getAllVmStats and list to allow a finer grained specification of data which should be included.
Parameter: statsType = <string> (getVmStats, getAllVmStats only) Allowed values:
* full (default to keep backwards compatibility) * app-list (Just send the application list) * rare (include everything from rarely changed to very frequent) * often (include everything from often changed to very frequent) * frequent (only send the very frequently changed items)
Parameter: clientId = <string> The client id is specified by the client and should be unique however constantly used.
Parameter: diff = <boolean> In combination with the clientId VDSM will send only differences to the previous request from the named clientId. (if diff=true)
Additional Change:
Besides the introduction of the new parameters for list, getVmStats and getAllVmStats it might make sense to include a hash for the appList into the rarely changed section of the response which would allow to identify changes and avoid having to sent the complete appList every so often and only if the hash known to the client is outdated.
Note: The appList (Application List) reported by the guest agent could be fully implemented on request only, as long as the guest agent installed supports this. As there seems to be a request to have the complete list of installed applications on all guests this data could be quite extensive and a huge list. On the other hand this data is only rarely visible and therefore it should not be requested all the time and only on demand. Improvement of the Guest Agent:
As part of the proposed solution it is necessary to improve the guest agent as well. For the full application list there should be implemented a caching system which will be fully reactive and should not poll the application list for example all the time. The guest can create a prepared data file containing all data in the JSON format (as used for the communication with VDSM via VIO) and just have to read that file from disk and directly sends it to VDSM. However it is quite possible that this list is to big and it might have to be chunked into pieces. (Multiple messages, which would have to be supported by VDSM then as well) The solution for this is to make VDSM request this data and it will retrieve the data necessary on request only. -- Regards,
Vinzenz Feenstra | Senior Software Engineer RedHat Engineering Virtualization R & D Phone: +420 532 294 625 IRC: vfeenstr or evilissimo
Better technology. Faster innovation. Powered by community collaboration. See how it works at redhat.com _______________________________________________ Engine-devel mailing list Engine-devel@ovirt.org http://lists.ovirt.org/mailman/listinfo/engine-devel
_______________________________________________ vdsm-devel mailing list vdsm-devel@lists.fedorahosted.org https://lists.fedorahosted.org/mailman/listinfo/vdsm-devel
----- Original Message -----
From: "Ayal Baron" <abaron@redhat.com> To: "Saggi Mizrahi" <smizrahi@redhat.com> Cc: engine-devel@ovirt.org, vdsm-devel@lists.fedorahosted.org, "Vinzenz Feenstra" <vfeenstr@redhat.com> Sent: Wednesday, March 13, 2013 5:39:24 PM Subject: Re: [vdsm] [Engine-devel] Proposal VDSM <=> Engine Data Statistics Retrieval Optimization
----- Original Message -----
I am completely against this. It make the return value differ according to input which is a big no no when talking about type safe APIs.
The only reason we have this problem is because there is this thing against making multiple calls.
Just split it up. getVmRuntimeStats() - transient things like mem and cpu% getVmInformation() - (semi)static things like disk\networking layout etc. Each updated at different intervals.
+1 on splitting the data up into 2 separate API calls. You could potentially add a checksum (md5, or any other way) of the "static" data to getVmRuntimeStats and not bother even with polling the VmInformation if this hasn't changed. Then you could poll as often as you'd like the stats and immediately see if you also need to retrieve VmInfo or not (you rarely would). +1 To Ayal's suggestion except that instead of the engine hashing the data VDSM sends the key which is opaque to the engine. This can be a local timestap or a generation number.
But, we might want to consider that when we add events polling becomes (much) less frequent so maybe it'll be an overkill.
----- Original Message -----
From: "Vinzenz Feenstra" <vfeenstr@redhat.com> To: vdsm-devel@lists.fedorahosted.org, engine-devel@ovirt.org Sent: Thursday, March 7, 2013 6:25:54 AM Subject: [Engine-devel] Proposal VDSM <=> Engine Data Statistics Retrieval Optimization
Please find the prettier version on the wiki: http://www.ovirt.org/Proposal_VDSM_-_Engine_Data_Statistics_Retrieval
Proposal VDSM - Engine Data Statistics Retrieval VDSM <=> Engine data retrieval optimization Motivation:
Currently the RHEVM engine is polling the a lot of data from VDSM every 15 seconds. This should be optimized and the amount of data requested should be more specific.
For each VM the data currently contains much more information than actually needed which blows up the size of the XML content quite big. We could optimize this by splitting the reply on the getVmStats based on the request of the engine into sections. For this reason Omer Frenkel and me have split up the data into parts based on their usage.
This data can and usually does change during the lifetime of the VM. Rarely Changed:
This data is change not very frequent and it should be enough to update this only once in a while. Most commonly this data changes after changes made in the UI or after a migration of the VM to another Host. Status = Running acpiEnable = true vmType = kvm guestName = W864GUESTAGENTT displayType = qxl guestOs = Win 8 kvmEnable = true # this should be constant and never changed pauseCode = NOERR monitorResponse = 0 session = Locked # unused netIfaces = [{'name': 'Realtek RTL8139C+ Fast Ethernet NIC', 'inet6': ['fe80::490c:92bb:bbcc:9f87'], 'inet': ['10.34.60.148'], 'hw': '00:1a:4a:22:3c:db'}] appsList = ['RHEV-Tools 3.2.4', 'RHEV-Agent64 3.2.3', 'RHEV-Serial64 3.2.3', 'RHEV-Network64 3.2.2', 'RHEV-Network64 3.2.3', 'RHEV-Block64 3.2.3', 'RHEV-Balloon64 3.2.3', 'RHEV-Balloon64 3.2.2', 'RHEV-Agent64 3.2.2', 'RHEV-USB 3.2.3', 'RHEV-Block64 3.2.2', 'RHEV-Serial64 3.2.2'] pid = 11314 guestIPs = 10.34.60.148 # duplicated info displayIp = 0 displayPort = 5902 displaySecurePort = 5903 username = user@W864GUESTAGENTT clientIp = lastLogin = 1361976900.67 Often Changed:
This data is changed quite often however it is not necessary to update this data every 15 seconds. As this is cumulative data and reflects the current status, and it does not need to be snapshotted every 15 seconds to retrieve statistics. The data can be retrieved in much more generous time slices. (e.g. Every 5 minutes) network = {'vnet1': {'macAddr': '00:1a:4a:22:3c:db', 'rxDropped': '0', 'txDropped': '0', 'rxErrors': '0', 'txRate': '0.0', 'rxRate': '0.0', 'txErrors': '0', 'state': 'unknown', 'speed': '100', 'name': 'vnet1'}} disksUsage = [{'path': 'c:\\', 'total': '64055406592', 'fs': 'NTFS', 'used': '19223846912'}, {'path': 'd:\\', 'total': '3490912256', 'fs': 'UDF', 'used': '3490912256'}] timeOffset = 14422 elapsedTime = 68591 hash = 2335461227228498964 statsAge = 0.09 # unused Often Changed but unused
This data does not seem to be used in the engine at all. It is not even used in the data warehouse. memoryStats = {'swap_out': '0', 'majflt': '0', 'mem_free': '1466884', 'swap_in': '0', 'pageflt': '0', 'mem_total': '2096736', 'mem_unused': '1466884'} balloonInfo = {'balloon_max': 2097152, 'balloon_cur': 2097152} disks = {'vda': {'readLatency': '0', 'apparentsize': '64424509440', 'writeLatency': '1754496', 'imageID': '28abb923-7b89-4638-84f8-1700f0b76482', 'flushLatency': '156549', 'readRate': '0.00', 'truesize': '18855059456', 'writeRate': '952.05'}, 'hdc': {'readLatency': '0', 'apparentsize': '0', 'writeLatency': '0', 'flushLatency': '0', 'readRate': '0.00', 'truesize': '0', 'writeRate': '0.00'}} Very frequent uppdates needed by webadmin portal:
This data is mostly needed for the webadmin portal and might be required to be updated quite often. An exception here is the statsAge field, which seems to be unused by the Engine. This data could be requested every 15 seconds to keep things as they are now. cpuSys = 2.32 cpuUser = 1.34 memUsage = 30 Proposed Solution for VDSM & Engine:
We will introduce new optional parameters to getVmStats, getAllVmStats and list to allow a finer grained specification of data which should be included.
Parameter: statsType = <string> (getVmStats, getAllVmStats only) Allowed values:
* full (default to keep backwards compatibility) * app-list (Just send the application list) * rare (include everything from rarely changed to very frequent) * often (include everything from often changed to very frequent) * frequent (only send the very frequently changed items)
Parameter: clientId = <string> The client id is specified by the client and should be unique however constantly used.
Parameter: diff = <boolean> In combination with the clientId VDSM will send only differences to the previous request from the named clientId. (if diff=true)
Additional Change:
Besides the introduction of the new parameters for list, getVmStats and getAllVmStats it might make sense to include a hash for the appList into the rarely changed section of the response which would allow to identify changes and avoid having to sent the complete appList every so often and only if the hash known to the client is outdated.
Note: The appList (Application List) reported by the guest agent could be fully implemented on request only, as long as the guest agent installed supports this. As there seems to be a request to have the complete list of installed applications on all guests this data could be quite extensive and a huge list. On the other hand this data is only rarely visible and therefore it should not be requested all the time and only on demand. Improvement of the Guest Agent:
As part of the proposed solution it is necessary to improve the guest agent as well. For the full application list there should be implemented a caching system which will be fully reactive and should not poll the application list for example all the time. The guest can create a prepared data file containing all data in the JSON format (as used for the communication with VDSM via VIO) and just have to read that file from disk and directly sends it to VDSM. However it is quite possible that this list is to big and it might have to be chunked into pieces. (Multiple messages, which would have to be supported by VDSM then as well) The solution for this is to make VDSM request this data and it will retrieve the data necessary on request only. -- Regards,
Vinzenz Feenstra | Senior Software Engineer RedHat Engineering Virtualization R & D Phone: +420 532 294 625 IRC: vfeenstr or evilissimo
Better technology. Faster innovation. Powered by community collaboration. See how it works at redhat.com _______________________________________________ Engine-devel mailing list Engine-devel@ovirt.org http://lists.ovirt.org/mailman/listinfo/engine-devel
_______________________________________________ vdsm-devel mailing list vdsm-devel@lists.fedorahosted.org https://lists.fedorahosted.org/mailman/listinfo/vdsm-devel
----- Original Message -----
----- Original Message -----
From: "Ayal Baron" <abaron@redhat.com> To: "Saggi Mizrahi" <smizrahi@redhat.com> Cc: engine-devel@ovirt.org, vdsm-devel@lists.fedorahosted.org, "Vinzenz Feenstra" <vfeenstr@redhat.com> Sent: Wednesday, March 13, 2013 5:39:24 PM Subject: Re: [vdsm] [Engine-devel] Proposal VDSM <=> Engine Data Statistics Retrieval Optimization
----- Original Message -----
I am completely against this. It make the return value differ according to input which is a big no no when talking about type safe APIs.
The only reason we have this problem is because there is this thing against making multiple calls.
Just split it up. getVmRuntimeStats() - transient things like mem and cpu% getVmInformation() - (semi)static things like disk\networking layout etc. Each updated at different intervals.
+1 on splitting the data up into 2 separate API calls. You could potentially add a checksum (md5, or any other way) of the "static" data to getVmRuntimeStats and not bother even with polling the VmInformation if this hasn't changed. Then you could poll as often as you'd like the stats and immediately see if you also need to retrieve VmInfo or not (you rarely would). +1 To Ayal's suggestion except that instead of the engine hashing the data VDSM sends the key which is opaque to the engine. This can be a local timestap or a generation number.
Of course vdsm does the hash, otherwise you'd need to pass all the data to engine which would beat the purpose.
But, we might want to consider that when we add events polling becomes (much) less frequent so maybe it'll be an overkill.
You'd still need to compare versions of the data in vdsm and send only if it changed. If you don't persist what was received last then potentially you could have a monday morning effect where upon on system startup you'd be sending everything. So I still think you'd want to have the hash.
----- Original Message -----
From: "Vinzenz Feenstra" <vfeenstr@redhat.com> To: vdsm-devel@lists.fedorahosted.org, engine-devel@ovirt.org Sent: Thursday, March 7, 2013 6:25:54 AM Subject: [Engine-devel] Proposal VDSM <=> Engine Data Statistics Retrieval Optimization
Please find the prettier version on the wiki: http://www.ovirt.org/Proposal_VDSM_-_Engine_Data_Statistics_Retrieval
Proposal VDSM - Engine Data Statistics Retrieval VDSM <=> Engine data retrieval optimization Motivation:
Currently the RHEVM engine is polling the a lot of data from VDSM every 15 seconds. This should be optimized and the amount of data requested should be more specific.
For each VM the data currently contains much more information than actually needed which blows up the size of the XML content quite big. We could optimize this by splitting the reply on the getVmStats based on the request of the engine into sections. For this reason Omer Frenkel and me have split up the data into parts based on their usage.
This data can and usually does change during the lifetime of the VM. Rarely Changed:
This data is change not very frequent and it should be enough to update this only once in a while. Most commonly this data changes after changes made in the UI or after a migration of the VM to another Host. Status = Running acpiEnable = true vmType = kvm guestName = W864GUESTAGENTT displayType = qxl guestOs = Win 8 kvmEnable = true # this should be constant and never changed pauseCode = NOERR monitorResponse = 0 session = Locked # unused netIfaces = [{'name': 'Realtek RTL8139C+ Fast Ethernet NIC', 'inet6': ['fe80::490c:92bb:bbcc:9f87'], 'inet': ['10.34.60.148'], 'hw': '00:1a:4a:22:3c:db'}] appsList = ['RHEV-Tools 3.2.4', 'RHEV-Agent64 3.2.3', 'RHEV-Serial64 3.2.3', 'RHEV-Network64 3.2.2', 'RHEV-Network64 3.2.3', 'RHEV-Block64 3.2.3', 'RHEV-Balloon64 3.2.3', 'RHEV-Balloon64 3.2.2', 'RHEV-Agent64 3.2.2', 'RHEV-USB 3.2.3', 'RHEV-Block64 3.2.2', 'RHEV-Serial64 3.2.2'] pid = 11314 guestIPs = 10.34.60.148 # duplicated info displayIp = 0 displayPort = 5902 displaySecurePort = 5903 username = user@W864GUESTAGENTT clientIp = lastLogin = 1361976900.67 Often Changed:
This data is changed quite often however it is not necessary to update this data every 15 seconds. As this is cumulative data and reflects the current status, and it does not need to be snapshotted every 15 seconds to retrieve statistics. The data can be retrieved in much more generous time slices. (e.g. Every 5 minutes) network = {'vnet1': {'macAddr': '00:1a:4a:22:3c:db', 'rxDropped': '0', 'txDropped': '0', 'rxErrors': '0', 'txRate': '0.0', 'rxRate': '0.0', 'txErrors': '0', 'state': 'unknown', 'speed': '100', 'name': 'vnet1'}} disksUsage = [{'path': 'c:\\', 'total': '64055406592', 'fs': 'NTFS', 'used': '19223846912'}, {'path': 'd:\\', 'total': '3490912256', 'fs': 'UDF', 'used': '3490912256'}] timeOffset = 14422 elapsedTime = 68591 hash = 2335461227228498964 statsAge = 0.09 # unused Often Changed but unused
This data does not seem to be used in the engine at all. It is not even used in the data warehouse. memoryStats = {'swap_out': '0', 'majflt': '0', 'mem_free': '1466884', 'swap_in': '0', 'pageflt': '0', 'mem_total': '2096736', 'mem_unused': '1466884'} balloonInfo = {'balloon_max': 2097152, 'balloon_cur': 2097152} disks = {'vda': {'readLatency': '0', 'apparentsize': '64424509440', 'writeLatency': '1754496', 'imageID': '28abb923-7b89-4638-84f8-1700f0b76482', 'flushLatency': '156549', 'readRate': '0.00', 'truesize': '18855059456', 'writeRate': '952.05'}, 'hdc': {'readLatency': '0', 'apparentsize': '0', 'writeLatency': '0', 'flushLatency': '0', 'readRate': '0.00', 'truesize': '0', 'writeRate': '0.00'}} Very frequent uppdates needed by webadmin portal:
This data is mostly needed for the webadmin portal and might be required to be updated quite often. An exception here is the statsAge field, which seems to be unused by the Engine. This data could be requested every 15 seconds to keep things as they are now. cpuSys = 2.32 cpuUser = 1.34 memUsage = 30 Proposed Solution for VDSM & Engine:
We will introduce new optional parameters to getVmStats, getAllVmStats and list to allow a finer grained specification of data which should be included.
Parameter: statsType = <string> (getVmStats, getAllVmStats only) Allowed values:
* full (default to keep backwards compatibility) * app-list (Just send the application list) * rare (include everything from rarely changed to very frequent) * often (include everything from often changed to very frequent) * frequent (only send the very frequently changed items)
Parameter: clientId = <string> The client id is specified by the client and should be unique however constantly used.
Parameter: diff = <boolean> In combination with the clientId VDSM will send only differences to the previous request from the named clientId. (if diff=true)
Additional Change:
Besides the introduction of the new parameters for list, getVmStats and getAllVmStats it might make sense to include a hash for the appList into the rarely changed section of the response which would allow to identify changes and avoid having to sent the complete appList every so often and only if the hash known to the client is outdated.
Note: The appList (Application List) reported by the guest agent could be fully implemented on request only, as long as the guest agent installed supports this. As there seems to be a request to have the complete list of installed applications on all guests this data could be quite extensive and a huge list. On the other hand this data is only rarely visible and therefore it should not be requested all the time and only on demand. Improvement of the Guest Agent:
As part of the proposed solution it is necessary to improve the guest agent as well. For the full application list there should be implemented a caching system which will be fully reactive and should not poll the application list for example all the time. The guest can create a prepared data file containing all data in the JSON format (as used for the communication with VDSM via VIO) and just have to read that file from disk and directly sends it to VDSM. However it is quite possible that this list is to big and it might have to be chunked into pieces. (Multiple messages, which would have to be supported by VDSM then as well) The solution for this is to make VDSM request this data and it will retrieve the data necessary on request only. -- Regards,
Vinzenz Feenstra | Senior Software Engineer RedHat Engineering Virtualization R & D Phone: +420 532 294 625 IRC: vfeenstr or evilissimo
Better technology. Faster innovation. Powered by community collaboration. See how it works at redhat.com _______________________________________________ Engine-devel mailing list Engine-devel@ovirt.org http://lists.ovirt.org/mailman/listinfo/engine-devel
_______________________________________________ vdsm-devel mailing list vdsm-devel@lists.fedorahosted.org https://lists.fedorahosted.org/mailman/listinfo/vdsm-devel
On 03/13/2013 11:55 PM, Ayal Baron wrote: ...
The only reason we have this problem is because there is this thing against making multiple calls.
Just split it up. getVmRuntimeStats() - transient things like mem and cpu% getVmInformation() - (semi)static things like disk\networking layout etc. Each updated at different intervals.
+1 on splitting the data up into 2 separate API calls. You could potentially add a checksum (md5, or any other way) of the "static" data to getVmRuntimeStats and not bother even with polling the VmInformation if this hasn't changed. Then you could poll as often as you'd like the stats and immediately see if you also need to retrieve VmInfo or not (you rarely would). +1 To Ayal's suggestion except that instead of the engine hashing the data VDSM sends the key which is opaque to the engine. This can be a local timestap or a generation number.
Of course vdsm does the hash, otherwise you'd need to pass all the data to engine which would beat the purpose.
I thought you meant engine will be sending the hash of previous requests per VM to vdsm, then vdsm will reply back with vm's removed, vm's added, and the details for vm's that changed (i.e., engine would be doing something like if-modified-since-checksum per vm). benefit is reducing a round trip. but first would need to split to calls of stats (always changing) and slowly/never changing data.
But, we might want to consider that when we add events polling becomes (much) less frequent so maybe it'll be an overkill.
You'd still need to compare versions of the data in vdsm and send only if it changed. If you don't persist what was received last then potentially you could have a monday morning effect where upon on system startup you'd be sending everything. So I still think you'd want to have the hash.
----- Original Message -----
From: "Vinzenz Feenstra" <vfeenstr@redhat.com> To: vdsm-devel@lists.fedorahosted.org, engine-devel@ovirt.org Sent: Thursday, March 7, 2013 6:25:54 AM Subject: [Engine-devel] Proposal VDSM <=> Engine Data Statistics Retrieval Optimization
Please find the prettier version on the wiki: http://www.ovirt.org/Proposal_VDSM_-_Engine_Data_Statistics_Retrieval
Proposal VDSM - Engine Data Statistics Retrieval VDSM <=> Engine data retrieval optimization Motivation:
Currently the RHEVM engine is polling the a lot of data from VDSM every 15 seconds. This should be optimized and the amount of data requested should be more specific.
For each VM the data currently contains much more information than actually needed which blows up the size of the XML content quite big. We could optimize this by splitting the reply on the getVmStats based on the request of the engine into sections. For this reason Omer Frenkel and me have split up the data into parts based on their usage.
This data can and usually does change during the lifetime of the VM. Rarely Changed:
This data is change not very frequent and it should be enough to update this only once in a while. Most commonly this data changes after changes made in the UI or after a migration of the VM to another Host. Status = Running acpiEnable = true vmType = kvm guestName = W864GUESTAGENTT displayType = qxl guestOs = Win 8 kvmEnable = true # this should be constant and never changed pauseCode = NOERR monitorResponse = 0 session = Locked # unused netIfaces = [{'name': 'Realtek RTL8139C+ Fast Ethernet NIC', 'inet6': ['fe80::490c:92bb:bbcc:9f87'], 'inet': ['10.34.60.148'], 'hw': '00:1a:4a:22:3c:db'}] appsList = ['RHEV-Tools 3.2.4', 'RHEV-Agent64 3.2.3', 'RHEV-Serial64 3.2.3', 'RHEV-Network64 3.2.2', 'RHEV-Network64 3.2.3', 'RHEV-Block64 3.2.3', 'RHEV-Balloon64 3.2.3', 'RHEV-Balloon64 3.2.2', 'RHEV-Agent64 3.2.2', 'RHEV-USB 3.2.3', 'RHEV-Block64 3.2.2', 'RHEV-Serial64 3.2.2'] pid = 11314 guestIPs = 10.34.60.148 # duplicated info displayIp = 0 displayPort = 5902 displaySecurePort = 5903 username = user@W864GUESTAGENTT clientIp = lastLogin = 1361976900.67 Often Changed:
This data is changed quite often however it is not necessary to update this data every 15 seconds. As this is cumulative data and reflects the current status, and it does not need to be snapshotted every 15 seconds to retrieve statistics. The data can be retrieved in much more generous time slices. (e.g. Every 5 minutes) network = {'vnet1': {'macAddr': '00:1a:4a:22:3c:db', 'rxDropped': '0', 'txDropped': '0', 'rxErrors': '0', 'txRate': '0.0', 'rxRate': '0.0', 'txErrors': '0', 'state': 'unknown', 'speed': '100', 'name': 'vnet1'}} disksUsage = [{'path': 'c:\\', 'total': '64055406592', 'fs': 'NTFS', 'used': '19223846912'}, {'path': 'd:\\', 'total': '3490912256', 'fs': 'UDF', 'used': '3490912256'}] timeOffset = 14422 elapsedTime = 68591 hash = 2335461227228498964 statsAge = 0.09 # unused Often Changed but unused
This data does not seem to be used in the engine at all. It is not even used in the data warehouse. memoryStats = {'swap_out': '0', 'majflt': '0', 'mem_free': '1466884', 'swap_in': '0', 'pageflt': '0', 'mem_total': '2096736', 'mem_unused': '1466884'} balloonInfo = {'balloon_max': 2097152, 'balloon_cur': 2097152} disks = {'vda': {'readLatency': '0', 'apparentsize': '64424509440', 'writeLatency': '1754496', 'imageID': '28abb923-7b89-4638-84f8-1700f0b76482', 'flushLatency': '156549', 'readRate': '0.00', 'truesize': '18855059456', 'writeRate': '952.05'}, 'hdc': {'readLatency': '0', 'apparentsize': '0', 'writeLatency': '0', 'flushLatency': '0', 'readRate': '0.00', 'truesize': '0', 'writeRate': '0.00'}} Very frequent uppdates needed by webadmin portal:
This data is mostly needed for the webadmin portal and might be required to be updated quite often. An exception here is the statsAge field, which seems to be unused by the Engine. This data could be requested every 15 seconds to keep things as they are now. cpuSys = 2.32 cpuUser = 1.34 memUsage = 30 Proposed Solution for VDSM & Engine:
We will introduce new optional parameters to getVmStats, getAllVmStats and list to allow a finer grained specification of data which should be included.
Parameter: statsType = <string> (getVmStats, getAllVmStats only) Allowed values:
* full (default to keep backwards compatibility) * app-list (Just send the application list) * rare (include everything from rarely changed to very frequent) * often (include everything from often changed to very frequent) * frequent (only send the very frequently changed items)
Parameter: clientId = <string> The client id is specified by the client and should be unique however constantly used.
Parameter: diff = <boolean> In combination with the clientId VDSM will send only differences to the previous request from the named clientId. (if diff=true)
Additional Change:
Besides the introduction of the new parameters for list, getVmStats and getAllVmStats it might make sense to include a hash for the appList into the rarely changed section of the response which would allow to identify changes and avoid having to sent the complete appList every so often and only if the hash known to the client is outdated.
Note: The appList (Application List) reported by the guest agent could be fully implemented on request only, as long as the guest agent installed supports this. As there seems to be a request to have the complete list of installed applications on all guests this data could be quite extensive and a huge list. On the other hand this data is only rarely visible and therefore it should not be requested all the time and only on demand. Improvement of the Guest Agent:
As part of the proposed solution it is necessary to improve the guest agent as well. For the full application list there should be implemented a caching system which will be fully reactive and should not poll the application list for example all the time. The guest can create a prepared data file containing all data in the JSON format (as used for the communication with VDSM via VIO) and just have to read that file from disk and directly sends it to VDSM. However it is quite possible that this list is to big and it might have to be chunked into pieces. (Multiple messages, which would have to be supported by VDSM then as well) The solution for this is to make VDSM request this data and it will retrieve the data necessary on request only. -- Regards,
Vinzenz Feenstra | Senior Software Engineer RedHat Engineering Virtualization R & D Phone: +420 532 294 625 IRC: vfeenstr or evilissimo
Better technology. Faster innovation. Powered by community collaboration. See how it works at redhat.com _______________________________________________ Engine-devel mailing list Engine-devel@ovirt.org http://lists.ovirt.org/mailman/listinfo/engine-devel
_______________________________________________ vdsm-devel mailing list vdsm-devel@lists.fedorahosted.org https://lists.fedorahosted.org/mailman/listinfo/vdsm-devel
_______________________________________________ vdsm-devel mailing list vdsm-devel@lists.fedorahosted.org https://lists.fedorahosted.org/mailman/listinfo/vdsm-devel
----- Original Message -----
On 03/13/2013 11:55 PM, Ayal Baron wrote: ...
The only reason we have this problem is because there is this thing against making multiple calls.
Just split it up. getVmRuntimeStats() - transient things like mem and cpu% getVmInformation() - (semi)static things like disk\networking layout etc. Each updated at different intervals.
+1 on splitting the data up into 2 separate API calls. You could potentially add a checksum (md5, or any other way) of the "static" data to getVmRuntimeStats and not bother even with polling the VmInformation if this hasn't changed. Then you could poll as often as you'd like the stats and immediately see if you also need to retrieve VmInfo or not (you rarely would). +1 To Ayal's suggestion except that instead of the engine hashing the data VDSM sends the key which is opaque to the engine. This can be a local timestap or a generation number.
Of course vdsm does the hash, otherwise you'd need to pass all the data to engine which would beat the purpose.
I thought you meant engine will be sending the hash of previous requests per VM to vdsm, then vdsm will reply back with vm's removed, vm's added, and the details for vm's that changed (i.e., engine would be doing something like if-modified-since-checksum per vm). benefit is reducing a round trip. but first would need to split to calls of stats (always changing) and slowly/never changing data.
If vdms accepts the hash then in your method engine would have to periodically call getVmInfo(hash). What I was suggesting is that getVmStats would return vmInfo hash so that we could avoid calling getVmInfo altogether. The stats *always* change so there is no need for checking if that info has changed. What we could do is avoid the split into 2 verbs by calling getVmStats(hash) and then have getVmStats return everything if the hash has changed or only the stats if it hasn't. This would be the least number of roundtrips and avoid the split. If you don't pass a hash it would return everything so this way it's also fully backward compatible.
But, we might want to consider that when we add events polling becomes (much) less frequent so maybe it'll be an overkill.
You'd still need to compare versions of the data in vdsm and send only if it changed. If you don't persist what was received last then potentially you could have a monday morning effect where upon on system startup you'd be sending everything. So I still think you'd want to have the hash.
----- Original Message -----
From: "Vinzenz Feenstra" <vfeenstr@redhat.com> To: vdsm-devel@lists.fedorahosted.org, engine-devel@ovirt.org Sent: Thursday, March 7, 2013 6:25:54 AM Subject: [Engine-devel] Proposal VDSM <=> Engine Data Statistics Retrieval Optimization
Please find the prettier version on the wiki: http://www.ovirt.org/Proposal_VDSM_-_Engine_Data_Statistics_Retrieval
Proposal VDSM - Engine Data Statistics Retrieval VDSM <=> Engine data retrieval optimization Motivation:
Currently the RHEVM engine is polling the a lot of data from VDSM every 15 seconds. This should be optimized and the amount of data requested should be more specific.
For each VM the data currently contains much more information than actually needed which blows up the size of the XML content quite big. We could optimize this by splitting the reply on the getVmStats based on the request of the engine into sections. For this reason Omer Frenkel and me have split up the data into parts based on their usage.
This data can and usually does change during the lifetime of the VM. Rarely Changed:
This data is change not very frequent and it should be enough to update this only once in a while. Most commonly this data changes after changes made in the UI or after a migration of the VM to another Host. Status = Running acpiEnable = true vmType = kvm guestName = W864GUESTAGENTT displayType = qxl guestOs = Win 8 kvmEnable = true # this should be constant and never changed pauseCode = NOERR monitorResponse = 0 session = Locked # unused netIfaces = [{'name': 'Realtek RTL8139C+ Fast Ethernet NIC', 'inet6': ['fe80::490c:92bb:bbcc:9f87'], 'inet': ['10.34.60.148'], 'hw': '00:1a:4a:22:3c:db'}] appsList = ['RHEV-Tools 3.2.4', 'RHEV-Agent64 3.2.3', 'RHEV-Serial64 3.2.3', 'RHEV-Network64 3.2.2', 'RHEV-Network64 3.2.3', 'RHEV-Block64 3.2.3', 'RHEV-Balloon64 3.2.3', 'RHEV-Balloon64 3.2.2', 'RHEV-Agent64 3.2.2', 'RHEV-USB 3.2.3', 'RHEV-Block64 3.2.2', 'RHEV-Serial64 3.2.2'] pid = 11314 guestIPs = 10.34.60.148 # duplicated info displayIp = 0 displayPort = 5902 displaySecurePort = 5903 username = user@W864GUESTAGENTT clientIp = lastLogin = 1361976900.67 Often Changed:
This data is changed quite often however it is not necessary to update this data every 15 seconds. As this is cumulative data and reflects the current status, and it does not need to be snapshotted every 15 seconds to retrieve statistics. The data can be retrieved in much more generous time slices. (e.g. Every 5 minutes) network = {'vnet1': {'macAddr': '00:1a:4a:22:3c:db', 'rxDropped': '0', 'txDropped': '0', 'rxErrors': '0', 'txRate': '0.0', 'rxRate': '0.0', 'txErrors': '0', 'state': 'unknown', 'speed': '100', 'name': 'vnet1'}} disksUsage = [{'path': 'c:\\', 'total': '64055406592', 'fs': 'NTFS', 'used': '19223846912'}, {'path': 'd:\\', 'total': '3490912256', 'fs': 'UDF', 'used': '3490912256'}] timeOffset = 14422 elapsedTime = 68591 hash = 2335461227228498964 statsAge = 0.09 # unused Often Changed but unused
This data does not seem to be used in the engine at all. It is not even used in the data warehouse. memoryStats = {'swap_out': '0', 'majflt': '0', 'mem_free': '1466884', 'swap_in': '0', 'pageflt': '0', 'mem_total': '2096736', 'mem_unused': '1466884'} balloonInfo = {'balloon_max': 2097152, 'balloon_cur': 2097152} disks = {'vda': {'readLatency': '0', 'apparentsize': '64424509440', 'writeLatency': '1754496', 'imageID': '28abb923-7b89-4638-84f8-1700f0b76482', 'flushLatency': '156549', 'readRate': '0.00', 'truesize': '18855059456', 'writeRate': '952.05'}, 'hdc': {'readLatency': '0', 'apparentsize': '0', 'writeLatency': '0', 'flushLatency': '0', 'readRate': '0.00', 'truesize': '0', 'writeRate': '0.00'}} Very frequent uppdates needed by webadmin portal:
This data is mostly needed for the webadmin portal and might be required to be updated quite often. An exception here is the statsAge field, which seems to be unused by the Engine. This data could be requested every 15 seconds to keep things as they are now. cpuSys = 2.32 cpuUser = 1.34 memUsage = 30 Proposed Solution for VDSM & Engine:
We will introduce new optional parameters to getVmStats, getAllVmStats and list to allow a finer grained specification of data which should be included.
Parameter: statsType = <string> (getVmStats, getAllVmStats only) Allowed values:
* full (default to keep backwards compatibility) * app-list (Just send the application list) * rare (include everything from rarely changed to very frequent) * often (include everything from often changed to very frequent) * frequent (only send the very frequently changed items)
Parameter: clientId = <string> The client id is specified by the client and should be unique however constantly used.
Parameter: diff = <boolean> In combination with the clientId VDSM will send only differences to the previous request from the named clientId. (if diff=true)
Additional Change:
Besides the introduction of the new parameters for list, getVmStats and getAllVmStats it might make sense to include a hash for the appList into the rarely changed section of the response which would allow to identify changes and avoid having to sent the complete appList every so often and only if the hash known to the client is outdated.
Note: The appList (Application List) reported by the guest agent could be fully implemented on request only, as long as the guest agent installed supports this. As there seems to be a request to have the complete list of installed applications on all guests this data could be quite extensive and a huge list. On the other hand this data is only rarely visible and therefore it should not be requested all the time and only on demand. Improvement of the Guest Agent:
As part of the proposed solution it is necessary to improve the guest agent as well. For the full application list there should be implemented a caching system which will be fully reactive and should not poll the application list for example all the time. The guest can create a prepared data file containing all data in the JSON format (as used for the communication with VDSM via VIO) and just have to read that file from disk and directly sends it to VDSM. However it is quite possible that this list is to big and it might have to be chunked into pieces. (Multiple messages, which would have to be supported by VDSM then as well) The solution for this is to make VDSM request this data and it will retrieve the data necessary on request only. -- Regards,
Vinzenz Feenstra | Senior Software Engineer RedHat Engineering Virtualization R & D Phone: +420 532 294 625 IRC: vfeenstr or evilissimo
Better technology. Faster innovation. Powered by community collaboration. See how it works at redhat.com _______________________________________________ Engine-devel mailing list Engine-devel@ovirt.org http://lists.ovirt.org/mailman/listinfo/engine-devel
_______________________________________________ vdsm-devel mailing list vdsm-devel@lists.fedorahosted.org https://lists.fedorahosted.org/mailman/listinfo/vdsm-devel
_______________________________________________ vdsm-devel mailing list vdsm-devel@lists.fedorahosted.org https://lists.fedorahosted.org/mailman/listinfo/vdsm-devel
On 17/03/13 15:13, Ayal Baron wrote:
----- Original Message -----
The only reason we have this problem is because there is this thing against making multiple calls.
Just split it up. getVmRuntimeStats() - transient things like mem and cpu% getVmInformation() - (semi)static things like disk\networking layout etc. Each updated at different intervals. +1 on splitting the data up into 2 separate API calls. You could potentially add a checksum (md5, or any other way) of the "static" data to getVmRuntimeStats and not bother even with polling the VmInformation if this hasn't changed. Then you could poll as often as you'd like the stats and immediately see if you also need to retrieve VmInfo or not (you rarely would). +1 To Ayal's suggestion except that instead of the engine hashing the data VDSM sends the key which is opaque to the engine. This can be a local timestap or a generation number. Of course vdsm does the hash, otherwise you'd need to pass all the data to engine which would beat the purpose. I thought you meant engine will be sending the hash of previous requests
On 03/13/2013 11:55 PM, Ayal Baron wrote: ... per VM to vdsm, then vdsm will reply back with vm's removed, vm's added, and the details for vm's that changed (i.e., engine would be doing something like if-modified-since-checksum per vm). benefit is reducing a round trip. but first would need to split to calls of stats (always changing) and slowly/never changing data. If vdms accepts the hash then in your method engine would have to periodically call getVmInfo(hash). What I was suggesting is that getVmStats would return vmInfo hash so that we could avoid calling getVmInfo altogether. The stats *always* change so there is no need for checking if that info has changed. What we could do is avoid the split into 2 verbs by calling getVmStats(hash) and then have getVmStats return everything if the hash has changed or only the stats if it hasn't. This would be the least number of roundtrips and avoid the split. If you don't pass a hash it would return everything so this way it's also fully backward compatible.
For the 'static' data, why is there a need for a hash? If VDSM sends in each update a timestamp, can't RHEVM just use if-modified-since with the last timestamp it got from VDSM? Is it cheaper for VDSM to calculate the hash, than update the timestamp per change in any of the fields? It doesn't really need to update the timestamp per change, only for the first change since last update sent actually (so 'dirty' flag in a way, to signify data that RHEVM hasn't seen yet). Y.
But, we might want to consider that when we add events polling becomes (much) less frequent so maybe it'll be an overkill. You'd still need to compare versions of the data in vdsm and send only if it changed. If you don't persist what was received last then potentially you could have a monday morning effect where upon on system startup you'd be sending everything. So I still think you'd want to have the hash.
----- Original Message ----- > From: "Vinzenz Feenstra" <vfeenstr@redhat.com> > To: vdsm-devel@lists.fedorahosted.org, engine-devel@ovirt.org > Sent: Thursday, March 7, 2013 6:25:54 AM > Subject: [Engine-devel] Proposal VDSM <=> Engine Data > Statistics > Retrieval Optimization > > > Please find the prettier version on the wiki: > http://www.ovirt.org/Proposal_VDSM_-_Engine_Data_Statistics_Retrieval > > Proposal VDSM - Engine Data Statistics Retrieval > VDSM <=> Engine data retrieval optimization > Motivation: > > > Currently the RHEVM engine is polling the a lot of data from > VDSM > every 15 seconds. This should be optimized and the amount of > data > requested should be more specific. > > For each VM the data currently contains much more information > than > actually needed which blows up the size of the XML content > quite > big. We could optimize this by splitting the reply on the > getVmStats > based on the request of the engine into sections. For this > reason > Omer Frenkel and me have split up the data into parts based on > their > usage. > > This data can and usually does change during the lifetime of > the > VM. > Rarely Changed: > > > This data is change not very frequent and it should be enough > to > update this only once in a while. Most commonly this data > changes > after changes made in the UI or after a migration of the VM to > another Host. Status = Running acpiEnable = true vmType = kvm > guestName = W864GUESTAGENTT displayType = qxl guestOs = Win 8 > kvmEnable = true # this should be constant and never changed > pauseCode = NOERR monitorResponse = 0 session = Locked # unused > netIfaces = [{'name': 'Realtek RTL8139C+ Fast Ethernet NIC', > 'inet6': ['fe80::490c:92bb:bbcc:9f87'], 'inet': > ['10.34.60.148'], > 'hw': '00:1a:4a:22:3c:db'}] appsList = ['RHEV-Tools 3.2.4', > 'RHEV-Agent64 3.2.3', 'RHEV-Serial64 3.2.3', 'RHEV-Network64 > 3.2.2', > 'RHEV-Network64 3.2.3', 'RHEV-Block64 3.2.3', 'RHEV-Balloon64 > 3.2.3', 'RHEV-Balloon64 3.2.2', 'RHEV-Agent64 3.2.2', 'RHEV-USB > 3.2.3', 'RHEV-Block64 3.2.2', 'RHEV-Serial64 3.2.2'] pid = > 11314 > guestIPs = 10.34.60.148 # duplicated info displayIp = 0 > displayPort > = 5902 displaySecurePort = 5903 username = user@W864GUESTAGENTT > clientIp = lastLogin = 1361976900.67 Often Changed: > > > This data is changed quite often however it is not necessary to > update this data every 15 seconds. As this is cumulative data > and > reflects the current status, and it does not need to be > snapshotted > every 15 seconds to retrieve statistics. The data can be > retrieved > in much more generous time slices. (e.g. Every 5 minutes) > network > = > {'vnet1': {'macAddr': '00:1a:4a:22:3c:db', 'rxDropped': '0', > 'txDropped': '0', 'rxErrors': '0', 'txRate': '0.0', 'rxRate': > '0.0', > 'txErrors': '0', 'state': 'unknown', 'speed': '100', 'name': > 'vnet1'}} disksUsage = [{'path': 'c:\\', 'total': > '64055406592', > 'fs': 'NTFS', 'used': '19223846912'}, {'path': 'd:\\', 'total': > '3490912256', 'fs': 'UDF', 'used': '3490912256'}] timeOffset = > 14422 > elapsedTime = 68591 hash = 2335461227228498964 statsAge = 0.09 > # > unused Often Changed but unused > > > This data does not seem to be used in the engine at all. It is > not > even used in the data warehouse. memoryStats = {'swap_out': > '0', > 'majflt': '0', 'mem_free': '1466884', 'swap_in': '0', > 'pageflt': > '0', 'mem_total': '2096736', 'mem_unused': '1466884'} > balloonInfo > = > {'balloon_max': 2097152, 'balloon_cur': 2097152} disks = > {'vda': > {'readLatency': '0', 'apparentsize': '64424509440', > 'writeLatency': > '1754496', 'imageID': '28abb923-7b89-4638-84f8-1700f0b76482', > 'flushLatency': '156549', 'readRate': '0.00', 'truesize': > '18855059456', 'writeRate': '952.05'}, 'hdc': {'readLatency': > '0', > 'apparentsize': '0', 'writeLatency': '0', 'flushLatency': '0', > 'readRate': '0.00', 'truesize': '0', 'writeRate': '0.00'}} Very > frequent uppdates needed by webadmin portal: > > > This data is mostly needed for the webadmin portal and might be > required to be updated quite often. An exception here is the > statsAge field, which seems to be unused by the Engine. This > data > could be requested every 15 seconds to keep things as they are > now. > cpuSys = 2.32 cpuUser = 1.34 memUsage = 30 Proposed Solution > for > VDSM & Engine: > > > We will introduce new optional parameters to getVmStats, > getAllVmStats and list to allow a finer grained specification > of > data which should be included. > > Parameter: statsType = <string> (getVmStats, getAllVmStats > only) > Allowed values: > > * full (default to keep backwards compatibility) > * app-list (Just send the application list) > * rare (include everything from rarely changed to very > frequent) > * often (include everything from often changed to very > frequent) > * frequent (only send the very frequently changed items) > > > > Parameter: clientId = <string> The client id is specified by > the > client and should be unique however constantly used. > > Parameter: diff = <boolean> In combination with the clientId > VDSM > will send only differences to the previous request from the > named > clientId. (if diff=true) > > > Additional Change: > > > Besides the introduction of the new parameters for list, > getVmStats > and getAllVmStats it might make sense to include a hash for the > appList into the rarely changed section of the response which > would > allow to identify changes and avoid having to sent the complete > appList every so often and only if the hash known to the client > is > outdated. > > Note: The appList (Application List) reported by the guest > agent > could be fully implemented on request only, as long as the > guest > agent installed supports this. As there seems to be a request > to > have the complete list of installed applications on all guests > this > data could be quite extensive and a huge list. On the other > hand > this data is only rarely visible and therefore it should not be > requested all the time and only on demand. Improvement of the > Guest > Agent: > > > As part of the proposed solution it is necessary to improve the > guest > agent as well. For the full application list there should be > implemented a caching system which will be fully reactive and > should > not poll the application list for example all the time. The > guest > can create a prepared data file containing all data in the JSON > format (as used for the communication with VDSM via VIO) and > just > have to read that file from disk and directly sends it to VDSM. > However it is quite possible that this list is to big and it > might > have to be chunked into pieces. (Multiple messages, which would > have > to be supported by VDSM then as well) The solution for this is > to > make VDSM request this data and it will retrieve the data > necessary > on request only. -- > Regards, > > Vinzenz Feenstra | Senior Software Engineer > RedHat Engineering Virtualization R & D > Phone: +420 532 294 625 > IRC: vfeenstr or evilissimo > > Better technology. Faster innovation. Powered by community > collaboration. > See how it works at redhat.com > _______________________________________________ > Engine-devel mailing list > Engine-devel@ovirt.org > http://lists.ovirt.org/mailman/listinfo/engine-devel > _______________________________________________ vdsm-devel mailing list vdsm-devel@lists.fedorahosted.org https://lists.fedorahosted.org/mailman/listinfo/vdsm-devel
vdsm-devel mailing list vdsm-devel@lists.fedorahosted.org https://lists.fedorahosted.org/mailman/listinfo/vdsm-devel
_______________________________________________ Engine-devel mailing list Engine-devel@ovirt.org http://lists.ovirt.org/mailman/listinfo/engine-devel
----- Original Message -----
On 17/03/13 15:13, Ayal Baron wrote:
> The only reason we have this problem is because there is this > thing against making multiple calls. > > Just split it up. > getVmRuntimeStats() - transient things like mem and cpu% > getVmInformation() - (semi)static things like disk\networking > layout > etc. > Each updated at different intervals. +1 on splitting the data up into 2 separate API calls. You could potentially add a checksum (md5, or any other way) of the "static" data to getVmRuntimeStats and not bother even with polling the VmInformation if this hasn't changed. Then you could poll as often as you'd like the stats and immediately see if you also need to retrieve VmInfo or not (you rarely would). +1 To Ayal's suggestion except that instead of the engine hashing the data VDSM sends the key which is opaque to the engine. This can be a local timestap or a generation number. Of course vdsm does the hash, otherwise you'd need to pass all the data to engine which would beat the purpose. I thought you meant engine will be sending the hash of previous requests
On 03/13/2013 11:55 PM, Ayal Baron wrote: ... per VM to vdsm, then vdsm will reply back with vm's removed, vm's added, and the details for vm's that changed (i.e., engine would be doing something like if-modified-since-checksum per vm). benefit is reducing a round trip. but first would need to split to calls of stats (always changing) and slowly/never changing data. If vdms accepts the hash then in your method engine would have to
----- Original Message ----- periodically call getVmInfo(hash). What I was suggesting is that getVmStats would return vmInfo hash so that we could avoid calling getVmInfo altogether. The stats *always* change so there is no need for checking if that info has changed. What we could do is avoid the split into 2 verbs by calling getVmStats(hash) and then have getVmStats return everything if the hash has changed or only the stats if it hasn't. This would be the least number of roundtrips and avoid the split. If you don't pass a hash it would return everything so this way it's also fully backward compatible.
For the 'static' data, why is there a need for a hash? If VDSM sends in each update a timestamp, can't RHEVM just use if-modified-since with the last timestamp it got from VDSM? Is it cheaper for VDSM to calculate the hash, than update the timestamp per change in any of the fields? It doesn't really need to update the timestamp per change, only for the first change since last update sent actually (so 'dirty' flag in a way, to signify data that RHEVM hasn't seen yet). Y.
As Saggi mentioned: "VDSM sends the key which is opaque to the engine. This can be a local timestap or a generation number." The content doesn't matter, what matters is that it has changed. timestamp assumes that vdsm will track changes and send only delta. Although possible this would be an overkill (for every value in the dict you'd have to hold a timestamp of last change and send only those which have changed since the timestamp which was passed by the user). Either way, I don't care what the 'hash' is, the point was that there is a simple way to keep a single API call, keep BC and toggle returning all data or just statistics (data that changes frequently) since last time user checked while minimizing API calls.
But, we might want to consider that when we add events polling becomes (much) less frequent so maybe it'll be an overkill. You'd still need to compare versions of the data in vdsm and send only if it changed. If you don't persist what was received last then potentially you could have a monday morning effect where upon on system startup you'd be sending everything. So I still think you'd want to have the hash.
> ----- Original Message ----- >> From: "Vinzenz Feenstra" <vfeenstr@redhat.com> >> To: vdsm-devel@lists.fedorahosted.org, engine-devel@ovirt.org >> Sent: Thursday, March 7, 2013 6:25:54 AM >> Subject: [Engine-devel] Proposal VDSM <=> Engine Data >> Statistics >> Retrieval Optimization >> >> >> Please find the prettier version on the wiki: >> http://www.ovirt.org/Proposal_VDSM_-_Engine_Data_Statistics_Retrieval >> >> Proposal VDSM - Engine Data Statistics Retrieval >> VDSM <=> Engine data retrieval optimization >> Motivation: >> >> >> Currently the RHEVM engine is polling the a lot of data from >> VDSM >> every 15 seconds. This should be optimized and the amount of >> data >> requested should be more specific. >> >> For each VM the data currently contains much more information >> than >> actually needed which blows up the size of the XML content >> quite >> big. We could optimize this by splitting the reply on the >> getVmStats >> based on the request of the engine into sections. For this >> reason >> Omer Frenkel and me have split up the data into parts based >> on >> their >> usage. >> >> This data can and usually does change during the lifetime of >> the >> VM. >> Rarely Changed: >> >> >> This data is change not very frequent and it should be enough >> to >> update this only once in a while. Most commonly this data >> changes >> after changes made in the UI or after a migration of the VM >> to >> another Host. Status = Running acpiEnable = true vmType = kvm >> guestName = W864GUESTAGENTT displayType = qxl guestOs = Win 8 >> kvmEnable = true # this should be constant and never changed >> pauseCode = NOERR monitorResponse = 0 session = Locked # >> unused >> netIfaces = [{'name': 'Realtek RTL8139C+ Fast Ethernet NIC', >> 'inet6': ['fe80::490c:92bb:bbcc:9f87'], 'inet': >> ['10.34.60.148'], >> 'hw': '00:1a:4a:22:3c:db'}] appsList = ['RHEV-Tools 3.2.4', >> 'RHEV-Agent64 3.2.3', 'RHEV-Serial64 3.2.3', 'RHEV-Network64 >> 3.2.2', >> 'RHEV-Network64 3.2.3', 'RHEV-Block64 3.2.3', 'RHEV-Balloon64 >> 3.2.3', 'RHEV-Balloon64 3.2.2', 'RHEV-Agent64 3.2.2', >> 'RHEV-USB >> 3.2.3', 'RHEV-Block64 3.2.2', 'RHEV-Serial64 3.2.2'] pid = >> 11314 >> guestIPs = 10.34.60.148 # duplicated info displayIp = 0 >> displayPort >> = 5902 displaySecurePort = 5903 username = >> user@W864GUESTAGENTT >> clientIp = lastLogin = 1361976900.67 Often Changed: >> >> >> This data is changed quite often however it is not necessary >> to >> update this data every 15 seconds. As this is cumulative data >> and >> reflects the current status, and it does not need to be >> snapshotted >> every 15 seconds to retrieve statistics. The data can be >> retrieved >> in much more generous time slices. (e.g. Every 5 minutes) >> network >> = >> {'vnet1': {'macAddr': '00:1a:4a:22:3c:db', 'rxDropped': '0', >> 'txDropped': '0', 'rxErrors': '0', 'txRate': '0.0', 'rxRate': >> '0.0', >> 'txErrors': '0', 'state': 'unknown', 'speed': '100', 'name': >> 'vnet1'}} disksUsage = [{'path': 'c:\\', 'total': >> '64055406592', >> 'fs': 'NTFS', 'used': '19223846912'}, {'path': 'd:\\', >> 'total': >> '3490912256', 'fs': 'UDF', 'used': '3490912256'}] timeOffset >> = >> 14422 >> elapsedTime = 68591 hash = 2335461227228498964 statsAge = >> 0.09 >> # >> unused Often Changed but unused >> >> >> This data does not seem to be used in the engine at all. It >> is >> not >> even used in the data warehouse. memoryStats = {'swap_out': >> '0', >> 'majflt': '0', 'mem_free': '1466884', 'swap_in': '0', >> 'pageflt': >> '0', 'mem_total': '2096736', 'mem_unused': '1466884'} >> balloonInfo >> = >> {'balloon_max': 2097152, 'balloon_cur': 2097152} disks = >> {'vda': >> {'readLatency': '0', 'apparentsize': '64424509440', >> 'writeLatency': >> '1754496', 'imageID': >> '28abb923-7b89-4638-84f8-1700f0b76482', >> 'flushLatency': '156549', 'readRate': '0.00', 'truesize': >> '18855059456', 'writeRate': '952.05'}, 'hdc': {'readLatency': >> '0', >> 'apparentsize': '0', 'writeLatency': '0', 'flushLatency': >> '0', >> 'readRate': '0.00', 'truesize': '0', 'writeRate': '0.00'}} >> Very >> frequent uppdates needed by webadmin portal: >> >> >> This data is mostly needed for the webadmin portal and might >> be >> required to be updated quite often. An exception here is the >> statsAge field, which seems to be unused by the Engine. This >> data >> could be requested every 15 seconds to keep things as they >> are >> now. >> cpuSys = 2.32 cpuUser = 1.34 memUsage = 30 Proposed Solution >> for >> VDSM & Engine: >> >> >> We will introduce new optional parameters to getVmStats, >> getAllVmStats and list to allow a finer grained specification >> of >> data which should be included. >> >> Parameter: statsType = <string> (getVmStats, getAllVmStats >> only) >> Allowed values: >> >> * full (default to keep backwards compatibility) >> * app-list (Just send the application list) >> * rare (include everything from rarely changed to very >> frequent) >> * often (include everything from often changed to very >> frequent) >> * frequent (only send the very frequently changed >> items) >> >> >> >> Parameter: clientId = <string> The client id is specified by >> the >> client and should be unique however constantly used. >> >> Parameter: diff = <boolean> In combination with the clientId >> VDSM >> will send only differences to the previous request from the >> named >> clientId. (if diff=true) >> >> >> Additional Change: >> >> >> Besides the introduction of the new parameters for list, >> getVmStats >> and getAllVmStats it might make sense to include a hash for >> the >> appList into the rarely changed section of the response which >> would >> allow to identify changes and avoid having to sent the >> complete >> appList every so often and only if the hash known to the >> client >> is >> outdated. >> >> Note: The appList (Application List) reported by the guest >> agent >> could be fully implemented on request only, as long as the >> guest >> agent installed supports this. As there seems to be a request >> to >> have the complete list of installed applications on all >> guests >> this >> data could be quite extensive and a huge list. On the other >> hand >> this data is only rarely visible and therefore it should not >> be >> requested all the time and only on demand. Improvement of the >> Guest >> Agent: >> >> >> As part of the proposed solution it is necessary to improve >> the >> guest >> agent as well. For the full application list there should be >> implemented a caching system which will be fully reactive and >> should >> not poll the application list for example all the time. The >> guest >> can create a prepared data file containing all data in the >> JSON >> format (as used for the communication with VDSM via VIO) and >> just >> have to read that file from disk and directly sends it to >> VDSM. >> However it is quite possible that this list is to big and it >> might >> have to be chunked into pieces. (Multiple messages, which >> would >> have >> to be supported by VDSM then as well) The solution for this >> is >> to >> make VDSM request this data and it will retrieve the data >> necessary >> on request only. -- >> Regards, >> >> Vinzenz Feenstra | Senior Software Engineer >> RedHat Engineering Virtualization R & D >> Phone: +420 532 294 625 >> IRC: vfeenstr or evilissimo >> >> Better technology. Faster innovation. Powered by community >> collaboration. >> See how it works at redhat.com >> _______________________________________________ >> Engine-devel mailing list >> Engine-devel@ovirt.org >> http://lists.ovirt.org/mailman/listinfo/engine-devel >> > _______________________________________________ > vdsm-devel mailing list > vdsm-devel@lists.fedorahosted.org > https://lists.fedorahosted.org/mailman/listinfo/vdsm-devel >
vdsm-devel mailing list vdsm-devel@lists.fedorahosted.org https://lists.fedorahosted.org/mailman/listinfo/vdsm-devel
_______________________________________________ Engine-devel mailing list Engine-devel@ovirt.org http://lists.ovirt.org/mailman/listinfo/engine-devel
On Sun, Mar 17, 2013 at 10:28:15AM -0400, Ayal Baron wrote:
----- Original Message -----
On 17/03/13 15:13, Ayal Baron wrote:
>> The only reason we have this problem is because there is this >> thing against making multiple calls. >> >> Just split it up. >> getVmRuntimeStats() - transient things like mem and cpu% >> getVmInformation() - (semi)static things like disk\networking >> layout >> etc. >> Each updated at different intervals. > +1 on splitting the data up into 2 separate API calls. > You could potentially add a checksum (md5, or any other way) of > the > "static" data to getVmRuntimeStats and not bother even with > polling > the VmInformation if this hasn't changed. Then you could poll > as > often as you'd like the stats and immediately see if you also > need > to retrieve VmInfo or not (you rarely would). +1 To Ayal's suggestion except that instead of the engine hashing the data VDSM sends the key which is opaque to the engine. This can be a local timestap or a generation number. Of course vdsm does the hash, otherwise you'd need to pass all the data to engine which would beat the purpose. I thought you meant engine will be sending the hash of previous requests
On 03/13/2013 11:55 PM, Ayal Baron wrote: ... per VM to vdsm, then vdsm will reply back with vm's removed, vm's added, and the details for vm's that changed (i.e., engine would be doing something like if-modified-since-checksum per vm). benefit is reducing a round trip. but first would need to split to calls of stats (always changing) and slowly/never changing data. If vdms accepts the hash then in your method engine would have to
----- Original Message ----- periodically call getVmInfo(hash). What I was suggesting is that getVmStats would return vmInfo hash so that we could avoid calling getVmInfo altogether. The stats *always* change so there is no need for checking if that info has changed. What we could do is avoid the split into 2 verbs by calling getVmStats(hash) and then have getVmStats return everything if the hash has changed or only the stats if it hasn't. This would be the least number of roundtrips and avoid the split. If you don't pass a hash it would return everything so this way it's also fully backward compatible.
For the 'static' data, why is there a need for a hash? If VDSM sends in each update a timestamp, can't RHEVM just use if-modified-since with the last timestamp it got from VDSM? Is it cheaper for VDSM to calculate the hash, than update the timestamp per change in any of the fields? It doesn't really need to update the timestamp per change, only for the first change since last update sent actually (so 'dirty' flag in a way, to signify data that RHEVM hasn't seen yet). Y.
As Saggi mentioned: "VDSM sends the key which is opaque to the engine. This can be a local timestap or a generation number."
The content doesn't matter, what matters is that it has changed. timestamp assumes that vdsm will track changes and send only delta. Although possible this would be an overkill (for every value in the dict you'd have to hold a timestamp of last change and send only those which have changed since the timestamp which was passed by the user).
If we're in the spirit of quoting Saggi, this suggestion is not compatible with "...mak[ing] the return value differ according to input ... is a big no no when talking about type safe APIs.". Dan.
This is a multi-part message in MIME format. --------------090700020909000108020704 Content-Type: text/plain; charset=UTF-8; format=flowed Content-Transfer-Encoding: 7bit Hi, I finally have an update for the proposal ready, after some thinking and throwing some ideas around with colleagues we came to the conclusion that the best approach might be as followed (without having dynamic return values) I have set up some test engine with 'FakeVDSM' and I am tracking the data sent. The system has 250 hosts and about 10000 VMs running. I have captured 1.3 GiB TCP data within about 20 minutes only from the communication between the engine and the fake VDSM host. In a frame of 23 minutes I have captured 13.560 getAllVMStats and getAllVdsStats calls, and 67827 calls to list. getAllVmStats replies have produced in that time about 510MiB of data transmitted to the engine. This proposal below would reduce the overall amount of data significantly and we're preparing to work on a prototype for the engine backend as well so we can backup this statement with hard numbers. Here's the proposal. VDSM <=> Engine data retrieval optimization Motivation: Currently the oVirt Engine is polling a lot of data from VDSM every 15 seconds. This should be optimized and the amount of data requested should be more specific. For each VM the data currently contains much more information than actually needed which blows up the size of the XML content quite big. We could optimize this by splitting the reply on the getVmStats based on the needs of the engine into different requests. For this reason Omer Frenkel and me have split up the data into parts based on their usage. Changes New Verbs getAllRuntimeStats Get runtime information of all VMs Returns for each VM a map with UUID and a value of: * *@cpuSys* Ratio of CPU time spent by qemu on other than guest time * *@cpuUser* Ratio of CPU time spent by the guest VM * *@memUsage* The percent of memory in use by the guest * *@elapsedTime* The number of seconds that the VM has been running * *@status* The current status of the given VM * *@statsAge* The age of these statistics in seconds * *@hashes* Hashes of several statistics and information around VMs Hashes consists of: * *@info* Hash for VmConfInfo data * *@config* Hash of the VM configuration XML * *@status* Hash of the VmStatusInfo data * *@guestDetails* Hash of the VmGuestDetails data getStatus Get status information about a list of VMs Parameters: * *@vmIDs* a list of UUIDs for VMs to query Returns for each VM in vmIDs a map with UUID and a value of: * *timeOffset* The time difference from host to the VM in seconds * *monitorResponse* Indicates if the qemu monitor is responsive * *clientIp* The IP address of the client connected to the display * *username* the username associated with the current session * *session* The current state of user interaction with the VM * *guestIPs* A space separated string of assigned IPv4 addresses * *pauseCode* Indicates the reason a VM has been paused getConfInfo Get configuration information about a list of VMs Parameters: * *@vmIDs* a list of UUIDs for VMs to query Returns for each VM in vmIDs a map with UUID and a value of: * *acpiEnable* Indicates if ACPI is enabled inside the VM * *displayPort* The port in use for unencrypted display data * *displaySecurePort* The port in use for encrypted display data * *displayType* The type of display in use * *displayIp* The IP address to use for accessing the VM display * *pid* The process ID of the underlying qemu process * *vmType* The type of VM * *kvmEnable* Indicates if KVM hardware acceleration is enabled * *cdrom* /*optional*/ The path to an ISO image used in the VM's CD-ROM device * *boot* /*optional*/ An alias for the type of device used to boot the VM getAllDeviceStats VM device statistics containing information for getting statistics and SLA information Returns for each VM a map with UUID and a value of: * *memoryStats* Memory statistics as reported by the guest agent * *balloonInfo* Guest memory balloon information * *disksUsage* Info about mounted filesystems as reported by the agent * *network* Network bandwidth/utilization statistics * *disks* Disk bandwidth/utilization statistics getGuestDetails Get details from the guest OS from a list of VMs Parameters: * *@vmIDs* a list of UUIDs for VMs to query Returns for each VM in vmIDs a map with UUID and a value of: * *appsList* A list of installed applications with their versions * *netIfaces* Network device address info as reported by the agent Usage Currently the engine is requesting currently every 3 seconds the vm list from each vdsm host and every 15 seconds all the data mentioned above for all VMs. The change would be as follows: The engine requests every 3 seconds getAllRuntimeStats from vdsm which will give the engine the most used data. If the engine has a mismatch of the hashes returned by getAllRuntimeStats it should request the data changed. if hashes.info changed => request getConfInfo with all vmIDs on that host where the hash changed if hashes.status changed => request getStatus with all vmIDs on that host where the hash changed if hashes.guestDetails changed => request getGuestDetails with all vmIDs on that host where the hash changed Request getAllDeviceStats periodically e.g. every 5 minutes, which should be sufficient for the DWH, in case it is not it could be even configurable. On 03/17/2013 04:30 PM, Dan Kenigsberg wrote:
On Sun, Mar 17, 2013 at 10:28:15AM -0400, Ayal Baron wrote:
----- Original Message -----
>>> The only reason we have this problem is because there is this >>> thing against making multiple calls. >>> >>> Just split it up. >>> getVmRuntimeStats() - transient things like mem and cpu% >>> getVmInformation() - (semi)static things like disk\networking >>> layout >>> etc. >>> Each updated at different intervals. >> +1 on splitting the data up into 2 separate API calls. >> You could potentially add a checksum (md5, or any other way) of >> the >> "static" data to getVmRuntimeStats and not bother even with >> polling >> the VmInformation if this hasn't changed. Then you could poll >> as >> often as you'd like the stats and immediately see if you also >> need >> to retrieve VmInfo or not (you rarely would). > +1 To Ayal's suggestion > except that instead of the engine hashing the data VDSM sends > the > key which is opaque to the engine. > This can be a local timestap or a generation number. Of course vdsm does the hash, otherwise you'd need to pass all the data to engine which would beat the purpose. I thought you meant engine will be sending the hash of previous requests
On 03/13/2013 11:55 PM, Ayal Baron wrote: ... per VM to vdsm, then vdsm will reply back with vm's removed, vm's added, and the details for vm's that changed (i.e., engine would be doing something like if-modified-since-checksum per vm). benefit is reducing a round trip. but first would need to split to calls of stats (always changing) and slowly/never changing data. If vdms accepts the hash then in your method engine would have to
----- Original Message ----- periodically call getVmInfo(hash). What I was suggesting is that getVmStats would return vmInfo hash so that we could avoid calling getVmInfo altogether. The stats *always* change so there is no need for checking if that info has changed. What we could do is avoid the split into 2 verbs by calling getVmStats(hash) and then have getVmStats return everything if the hash has changed or only the stats if it hasn't. This would be the least number of roundtrips and avoid the split. If you don't pass a hash it would return everything so this way it's also fully backward compatible. For the 'static' data, why is there a need for a hash? If VDSM sends in each update a timestamp, can't RHEVM just use if-modified-since with the last timestamp it got from VDSM? Is it cheaper for VDSM to calculate the hash, than update the timestamp
On 17/03/13 15:13, Ayal Baron wrote: per change in any of the fields? It doesn't really need to update the timestamp per change, only for the first change since last update sent actually (so 'dirty' flag in a way, to signify data that RHEVM hasn't seen yet). Y. As Saggi mentioned: "VDSM sends the key which is opaque to the engine. This can be a local timestap or a generation number."
The content doesn't matter, what matters is that it has changed. timestamp assumes that vdsm will track changes and send only delta. Although possible this would be an overkill (for every value in the dict you'd have to hold a timestamp of last change and send only those which have changed since the timestamp which was passed by the user).
If we're in the spirit of quoting Saggi, this suggestion is not compatible with "...mak[ing] the return value differ according to input ... is a big no no when talking about type safe APIs.".
Dan. _______________________________________________ vdsm-devel mailing list vdsm-devel@lists.fedorahosted.org https://lists.fedorahosted.org/mailman/listinfo/vdsm-devel
-- Regards, Vinzenz Feenstra | Senior Software Engineer RedHat Engineering Virtualization R & D Phone: +420 532 294 625 IRC: vfeenstr or evilissimo Better technology. Faster innovation. Powered by community collaboration. See how it works at redhat.com --------------090700020909000108020704 Content-Type: text/html; charset=UTF-8 Content-Transfer-Encoding: 7bit <html> <head> <meta content="text/html; charset=UTF-8" http-equiv="Content-Type"> </head> <body bgcolor="#FFFFFF" text="#000000"> <div class="moz-cite-prefix">Hi,<br> <br> I finally have an update for the proposal ready, after some thinking and throwing some ideas around with colleagues we came to the conclusion that the best approach might be as followed (without having dynamic return values)<br> <br> I have set up some test engine with 'FakeVDSM' and I am tracking the data sent. The system has 250 hosts and about 10000 VMs running. I have captured 1.3 GiB TCP data within about 20 minutes only from the communication between the engine and the fake VDSM host. In a frame of 23 minutes I have captured <meta http-equiv="CONTENT-TYPE" content="text/html; charset=UTF-8"> <title></title> <meta name="GENERATOR" content="LibreOffice 3.6 (Linux)"> <style> <!-- BODY,DIV,TABLE,THEAD,TBODY,TFOOT,TR,TH,TD,P { font-family:"Liberation Sans"; font-size:x-small } --> </style>13.560 getAllVMStats and getAllVdsStats calls, and 67827 calls to list.<br> <br> getAllVmStats replies have produced in that time about 510MiB of data transmitted to the engine. This proposal below would reduce the overall amount of data significantly and we're preparing to work on a prototype for the engine backend as well so we can backup this statement with hard numbers.<br> <br> <br> Here's the proposal.<br> <br> <h2><span class="mw-headline" id="VDSM_.3C.3D.3E_Engine_data_retrieval_optimization">VDSM <=> Engine data retrieval optimization </span></h2> <h3> <span class="mw-headline" id="Motivation:"> Motivation: </span></h3> <p>Currently the oVirt Engine is polling a lot of data from VDSM every 15 seconds. This should be optimized and the amount of data requested should be more specific. </p> <p>For each VM the data currently contains much more information than actually needed which blows up the size of the XML content quite big. We could optimize this by splitting the reply on the getVmStats based on the needs of the engine into different requests. For this reason Omer Frenkel and me have split up the data into parts based on their usage. </p> <h3> <span class="mw-headline" id="Changes"> Changes </span></h3> <h4> <span class="mw-headline" id="New_Verbs"> New Verbs </span></h4> <h5> <span class="mw-headline" id="getAllRuntimeStats"> getAllRuntimeStats </span></h5> <p>Get runtime information of all VMs<br> Returns for each VM a map with UUID and a value of: </p> <ul> <li> <b>@cpuSys</b> Ratio of CPU time spent by qemu on other than guest time </li> <li> <b>@cpuUser</b> Ratio of CPU time spent by the guest VM </li> <li> <b>@memUsage</b> The percent of memory in use by the guest </li> <li> <b>@elapsedTime</b> The number of seconds that the VM has been running </li> <li> <b>@status</b> The current status of the given VM </li> <li> <b>@statsAge</b> The age of these statistics in seconds </li> <li> <b>@hashes</b> Hashes of several statistics and information around VMs </li> </ul> <p>Hashes consists of: </p> <ul> <li> <b>@info</b> Hash for VmConfInfo data </li> <li> <b>@config</b> Hash of the VM configuration XML </li> <li> <b>@status</b> Hash of the VmStatusInfo data </li> <li> <b>@guestDetails</b> Hash of the VmGuestDetails data </li> </ul> <h5> <span class="mw-headline" id="getStatus"> getStatus </span></h5> <p>Get status information about a list of VMs<br> Parameters: </p> <ul> <li> <b>@vmIDs</b> a list of UUIDs for VMs to query </li> </ul> <p>Returns for each VM in vmIDs a map with UUID and a value of: </p> <ul> <li> <b>timeOffset</b> The time difference from host to the VM in seconds </li> <li> <b>monitorResponse</b> Indicates if the qemu monitor is responsive </li> <li> <b>clientIp</b> The IP address of the client connected to the display </li> <li> <b>username</b> the username associated with the current session </li> <li> <b>session</b> The current state of user interaction with the VM </li> <li> <b>guestIPs</b> A space separated string of assigned IPv4 addresses </li> <li> <b>pauseCode</b> Indicates the reason a VM has been paused </li> </ul> <h5> <span class="mw-headline" id="getConfInfo"> getConfInfo </span></h5> <p>Get configuration information about a list of VMs<br> Parameters: </p> <ul> <li> <b>@vmIDs</b> a list of UUIDs for VMs to query </li> </ul> <p>Returns for each VM in vmIDs a map with UUID and a value of: </p> <ul> <li> <b>acpiEnable</b> Indicates if ACPI is enabled inside the VM </li> <li> <b>displayPort</b> The port in use for unencrypted display data </li> <li> <b>displaySecurePort</b> The port in use for encrypted display data </li> <li> <b>displayType</b> The type of display in use </li> <li> <b>displayIp</b> The IP address to use for accessing the VM display </li> <li> <b>pid</b> The process ID of the underlying qemu process </li> <li> <b>vmType</b> The type of VM </li> <li> <b>kvmEnable</b> Indicates if KVM hardware acceleration is enabled </li> <li> <b>cdrom</b> <i><b>optional</b></i> The path to an ISO image used in the VM's CD-ROM device </li> <li> <b>boot</b> <i><b>optional</b></i> An alias for the type of device used to boot the VM </li> </ul> <h5> <span class="mw-headline" id="getAllDeviceStats"> getAllDeviceStats </span></h5> <p>VM device statistics containing information for getting statistics and SLA information<br> Returns for each VM a map with UUID and a value of: </p> <ul> <li> <b>memoryStats</b> Memory statistics as reported by the guest agent </li> <li> <b>balloonInfo</b> Guest memory balloon information </li> <li> <b>disksUsage</b> Info about mounted filesystems as reported by the agent </li> <li> <b>network</b> Network bandwidth/utilization statistics </li> <li> <b>disks</b> Disk bandwidth/utilization statistics </li> </ul> <h5> <span class="mw-headline" id="getGuestDetails"> getGuestDetails </span></h5> <p>Get details from the guest OS from a list of VMs<br> Parameters: </p> <ul> <li> <b>@vmIDs</b> a list of UUIDs for VMs to query </li> </ul> <p>Returns for each VM in vmIDs a map with UUID and a value of: </p> <ul> <li> <b>appsList</b> A list of installed applications with their versions </li> <li> <b>netIfaces</b> Network device address info as reported by the agent </li> </ul> <h4> <span class="mw-headline" id="Usage"> Usage </span></h4> <p>Currently the engine is requesting currently every 3 seconds the vm list from each vdsm host and every 15 seconds all the data mentioned above for all VMs. </p> <p>The change would be as follows: </p> <p>The engine requests every 3 seconds getAllRuntimeStats from vdsm which will give the engine the most used data. If the engine has a mismatch of the hashes returned by getAllRuntimeStats it should request the data changed. </p> <p>if hashes.info changed => request getConfInfo with all vmIDs on that host where the hash changed if hashes.status changed => request getStatus with all vmIDs on that host where the hash changed if hashes.guestDetails changed => request getGuestDetails with all vmIDs on that host where the hash changed </p> <p>Request getAllDeviceStats periodically e.g. every 5 minutes, which should be sufficient for the DWH, in case it is not it could be even configurable. </p> <br> <br> On 03/17/2013 04:30 PM, Dan Kenigsberg wrote:<br> </div> <blockquote cite="mid:20130317153038.GB4891@redhat.com" type="cite"> <pre wrap="">On Sun, Mar 17, 2013 at 10:28:15AM -0400, Ayal Baron wrote: </pre> <blockquote type="cite"> <pre wrap=""> ----- Original Message ----- </pre> <blockquote type="cite"> <pre wrap="">On 17/03/13 15:13, Ayal Baron wrote: </pre> <blockquote type="cite"> <pre wrap=""> ----- Original Message ----- </pre> <blockquote type="cite"> <pre wrap="">On 03/13/2013 11:55 PM, Ayal Baron wrote: ... </pre> <blockquote type="cite"> <blockquote type="cite"> <blockquote type="cite"> <blockquote type="cite"> <pre wrap="">The only reason we have this problem is because there is this thing against making multiple calls. Just split it up. getVmRuntimeStats() - transient things like mem and cpu% getVmInformation() - (semi)static things like disk\networking layout etc. Each updated at different intervals. </pre> </blockquote> <pre wrap="">+1 on splitting the data up into 2 separate API calls. You could potentially add a checksum (md5, or any other way) of the "static" data to getVmRuntimeStats and not bother even with polling the VmInformation if this hasn't changed. Then you could poll as often as you'd like the stats and immediately see if you also need to retrieve VmInfo or not (you rarely would). </pre> </blockquote> <pre wrap="">+1 To Ayal's suggestion except that instead of the engine hashing the data VDSM sends the key which is opaque to the engine. This can be a local timestap or a generation number. </pre> </blockquote> <pre wrap="">Of course vdsm does the hash, otherwise you'd need to pass all the data to engine which would beat the purpose. </pre> </blockquote> <pre wrap="">I thought you meant engine will be sending the hash of previous requests per VM to vdsm, then vdsm will reply back with vm's removed, vm's added, and the details for vm's that changed (i.e., engine would be doing something like if-modified-since-checksum per vm). benefit is reducing a round trip. but first would need to split to calls of stats (always changing) and slowly/never changing data. </pre> </blockquote> <pre wrap="">If vdms accepts the hash then in your method engine would have to periodically call getVmInfo(hash). What I was suggesting is that getVmStats would return vmInfo hash so that we could avoid calling getVmInfo altogether. The stats *always* change so there is no need for checking if that info has changed. What we could do is avoid the split into 2 verbs by calling getVmStats(hash) and then have getVmStats return everything if the hash has changed or only the stats if it hasn't. This would be the least number of roundtrips and avoid the split. If you don't pass a hash it would return everything so this way it's also fully backward compatible. </pre> </blockquote> <pre wrap=""> For the 'static' data, why is there a need for a hash? If VDSM sends in each update a timestamp, can't RHEVM just use if-modified-since with the last timestamp it got from VDSM? Is it cheaper for VDSM to calculate the hash, than update the timestamp per change in any of the fields? It doesn't really need to update the timestamp per change, only for the first change since last update sent actually (so 'dirty' flag in a way, to signify data that RHEVM hasn't seen yet). Y. </pre> </blockquote> <pre wrap=""> As Saggi mentioned: "VDSM sends the key which is opaque to the engine. This can be a local timestap or a generation number." The content doesn't matter, what matters is that it has changed. timestamp assumes that vdsm will track changes and send only delta. Although possible this would be an overkill (for every value in the dict you'd have to hold a timestamp of last change and send only those which have changed since the timestamp which was passed by the user). </pre> </blockquote> <pre wrap=""> If we're in the spirit of quoting Saggi, this suggestion is not compatible with "...mak[ing] the return value differ according to input ... is a big no no when talking about type safe APIs.". Dan. _______________________________________________ vdsm-devel mailing list <a class="moz-txt-link-abbreviated" href="mailto:vdsm-devel@lists.fedorahosted.org">vdsm-devel@lists.fedorahosted.org</a> <a class="moz-txt-link-freetext" href="https://lists.fedorahosted.org/mailman/listinfo/vdsm-devel">https://lists.fedorahosted.org/mailman/listinfo/vdsm-devel</a> </pre> </blockquote> <br> <br> <pre class="moz-signature" cols="72">-- Regards, Vinzenz Feenstra | Senior Software Engineer RedHat Engineering Virtualization R & D Phone: +420 532 294 625 IRC: vfeenstr or evilissimo Better technology. Faster innovation. Powered by community collaboration. See how it works at redhat.com</pre> </body> </html> --------------090700020909000108020704--
----- Original Message -----
From: "Ayal Baron" <abaron@redhat.com> To: "Itamar Heim" <iheim@redhat.com> Cc: engine-devel@ovirt.org, vdsm-devel@lists.fedorahosted.org Sent: Sunday, March 17, 2013 3:13:09 PM Subject: Re: [Engine-devel] [vdsm] Proposal VDSM <=> Engine Data Statistics Retrieval Optimization
----- Original Message -----
On 03/13/2013 11:55 PM, Ayal Baron wrote: ...
The only reason we have this problem is because there is this thing against making multiple calls.
Just split it up. getVmRuntimeStats() - transient things like mem and cpu% getVmInformation() - (semi)static things like disk\networking layout etc. Each updated at different intervals.
+1 on splitting the data up into 2 separate API calls. You could potentially add a checksum (md5, or any other way) of the "static" data to getVmRuntimeStats and not bother even with polling the VmInformation if this hasn't changed. Then you could poll as often as you'd like the stats and immediately see if you also need to retrieve VmInfo or not (you rarely would). +1 To Ayal's suggestion except that instead of the engine hashing the data VDSM sends the key which is opaque to the engine. This can be a local timestap or a generation number.
Of course vdsm does the hash, otherwise you'd need to pass all the data to engine which would beat the purpose.
I thought you meant engine will be sending the hash of previous requests per VM to vdsm, then vdsm will reply back with vm's removed, vm's added, and the details for vm's that changed (i.e., engine would be doing something like if-modified-since-checksum per vm). benefit is reducing a round trip. but first would need to split to calls of stats (always changing) and slowly/never changing data.
If vdms accepts the hash then in your method engine would have to periodically call getVmInfo(hash). What I was suggesting is that getVmStats would return vmInfo hash so that we could avoid calling getVmInfo altogether. The stats *always* change so there is no need for checking if that info has changed. What we could do is avoid the split into 2 verbs by calling getVmStats(hash) and then have getVmStats return everything if the hash has changed or only the stats if it hasn't. This would be the least number of roundtrips and avoid the split. If you don't pass a hash it would return everything so this way it's also fully backward compatible.
Actually, I assume we can pass hash 0 (to have vdsm return "everything"). I assume that the chances for md5 on "real data" (i.e - real data that is known to engine) to be 0 are very slim.
But, we might want to consider that when we add events polling becomes (much) less frequent so maybe it'll be an overkill.
You'd still need to compare versions of the data in vdsm and send only if it changed. If you don't persist what was received last then potentially you could have a monday morning effect where upon on system startup you'd be sending everything. So I still think you'd want to have the hash.
----- Original Message ----- > From: "Vinzenz Feenstra" <vfeenstr@redhat.com> > To: vdsm-devel@lists.fedorahosted.org, engine-devel@ovirt.org > Sent: Thursday, March 7, 2013 6:25:54 AM > Subject: [Engine-devel] Proposal VDSM <=> Engine Data > Statistics > Retrieval Optimization > > > Please find the prettier version on the wiki: > http://www.ovirt.org/Proposal_VDSM_-_Engine_Data_Statistics_Retrieval > > Proposal VDSM - Engine Data Statistics Retrieval > VDSM <=> Engine data retrieval optimization > Motivation: > > > Currently the RHEVM engine is polling the a lot of data from > VDSM > every 15 seconds. This should be optimized and the amount of > data > requested should be more specific. > > For each VM the data currently contains much more information > than > actually needed which blows up the size of the XML content > quite > big. We could optimize this by splitting the reply on the > getVmStats > based on the request of the engine into sections. For this > reason > Omer Frenkel and me have split up the data into parts based > on > their > usage. > > This data can and usually does change during the lifetime of > the > VM. > Rarely Changed: > > > This data is change not very frequent and it should be enough > to > update this only once in a while. Most commonly this data > changes > after changes made in the UI or after a migration of the VM > to > another Host. Status = Running acpiEnable = true vmType = kvm > guestName = W864GUESTAGENTT displayType = qxl guestOs = Win 8 > kvmEnable = true # this should be constant and never changed > pauseCode = NOERR monitorResponse = 0 session = Locked # > unused > netIfaces = [{'name': 'Realtek RTL8139C+ Fast Ethernet NIC', > 'inet6': ['fe80::490c:92bb:bbcc:9f87'], 'inet': > ['10.34.60.148'], > 'hw': '00:1a:4a:22:3c:db'}] appsList = ['RHEV-Tools 3.2.4', > 'RHEV-Agent64 3.2.3', 'RHEV-Serial64 3.2.3', 'RHEV-Network64 > 3.2.2', > 'RHEV-Network64 3.2.3', 'RHEV-Block64 3.2.3', 'RHEV-Balloon64 > 3.2.3', 'RHEV-Balloon64 3.2.2', 'RHEV-Agent64 3.2.2', > 'RHEV-USB > 3.2.3', 'RHEV-Block64 3.2.2', 'RHEV-Serial64 3.2.2'] pid = > 11314 > guestIPs = 10.34.60.148 # duplicated info displayIp = 0 > displayPort > = 5902 displaySecurePort = 5903 username = > user@W864GUESTAGENTT > clientIp = lastLogin = 1361976900.67 Often Changed: > > > This data is changed quite often however it is not necessary > to > update this data every 15 seconds. As this is cumulative data > and > reflects the current status, and it does not need to be > snapshotted > every 15 seconds to retrieve statistics. The data can be > retrieved > in much more generous time slices. (e.g. Every 5 minutes) > network > = > {'vnet1': {'macAddr': '00:1a:4a:22:3c:db', 'rxDropped': '0', > 'txDropped': '0', 'rxErrors': '0', 'txRate': '0.0', 'rxRate': > '0.0', > 'txErrors': '0', 'state': 'unknown', 'speed': '100', 'name': > 'vnet1'}} disksUsage = [{'path': 'c:\\', 'total': > '64055406592', > 'fs': 'NTFS', 'used': '19223846912'}, {'path': 'd:\\', > 'total': > '3490912256', 'fs': 'UDF', 'used': '3490912256'}] timeOffset > = > 14422 > elapsedTime = 68591 hash = 2335461227228498964 statsAge = > 0.09 > # > unused Often Changed but unused > > > This data does not seem to be used in the engine at all. It > is > not > even used in the data warehouse. memoryStats = {'swap_out': > '0', > 'majflt': '0', 'mem_free': '1466884', 'swap_in': '0', > 'pageflt': > '0', 'mem_total': '2096736', 'mem_unused': '1466884'} > balloonInfo > = > {'balloon_max': 2097152, 'balloon_cur': 2097152} disks = > {'vda': > {'readLatency': '0', 'apparentsize': '64424509440', > 'writeLatency': > '1754496', 'imageID': > '28abb923-7b89-4638-84f8-1700f0b76482', > 'flushLatency': '156549', 'readRate': '0.00', 'truesize': > '18855059456', 'writeRate': '952.05'}, 'hdc': {'readLatency': > '0', > 'apparentsize': '0', 'writeLatency': '0', 'flushLatency': > '0', > 'readRate': '0.00', 'truesize': '0', 'writeRate': '0.00'}} > Very > frequent uppdates needed by webadmin portal: > > > This data is mostly needed for the webadmin portal and might > be > required to be updated quite often. An exception here is the > statsAge field, which seems to be unused by the Engine. This > data > could be requested every 15 seconds to keep things as they > are > now. > cpuSys = 2.32 cpuUser = 1.34 memUsage = 30 Proposed Solution > for > VDSM & Engine: > > > We will introduce new optional parameters to getVmStats, > getAllVmStats and list to allow a finer grained specification > of > data which should be included. > > Parameter: statsType = <string> (getVmStats, getAllVmStats > only) > Allowed values: > > * full (default to keep backwards compatibility) > * app-list (Just send the application list) > * rare (include everything from rarely changed to very > frequent) > * often (include everything from often changed to very > frequent) > * frequent (only send the very frequently changed items) > > > > Parameter: clientId = <string> The client id is specified by > the > client and should be unique however constantly used. > > Parameter: diff = <boolean> In combination with the clientId > VDSM > will send only differences to the previous request from the > named > clientId. (if diff=true) > > > Additional Change: > > > Besides the introduction of the new parameters for list, > getVmStats > and getAllVmStats it might make sense to include a hash for > the > appList into the rarely changed section of the response which > would > allow to identify changes and avoid having to sent the > complete > appList every so often and only if the hash known to the > client > is > outdated. > > Note: The appList (Application List) reported by the guest > agent > could be fully implemented on request only, as long as the > guest > agent installed supports this. As there seems to be a request > to > have the complete list of installed applications on all > guests > this > data could be quite extensive and a huge list. On the other > hand > this data is only rarely visible and therefore it should not > be > requested all the time and only on demand. Improvement of the > Guest > Agent: > > > As part of the proposed solution it is necessary to improve > the > guest > agent as well. For the full application list there should be > implemented a caching system which will be fully reactive and > should > not poll the application list for example all the time. The > guest > can create a prepared data file containing all data in the > JSON > format (as used for the communication with VDSM via VIO) and > just > have to read that file from disk and directly sends it to > VDSM. > However it is quite possible that this list is to big and it > might > have to be chunked into pieces. (Multiple messages, which > would > have > to be supported by VDSM then as well) The solution for this > is > to > make VDSM request this data and it will retrieve the data > necessary > on request only. -- > Regards, > > Vinzenz Feenstra | Senior Software Engineer > RedHat Engineering Virtualization R & D > Phone: +420 532 294 625 > IRC: vfeenstr or evilissimo > > Better technology. Faster innovation. Powered by community > collaboration. > See how it works at redhat.com > _______________________________________________ > Engine-devel mailing list > Engine-devel@ovirt.org > http://lists.ovirt.org/mailman/listinfo/engine-devel > _______________________________________________ vdsm-devel mailing list vdsm-devel@lists.fedorahosted.org https://lists.fedorahosted.org/mailman/listinfo/vdsm-devel
_______________________________________________ vdsm-devel mailing list vdsm-devel@lists.fedorahosted.org https://lists.fedorahosted.org/mailman/listinfo/vdsm-devel
_______________________________________________ Engine-devel mailing list Engine-devel@ovirt.org http://lists.ovirt.org/mailman/listinfo/engine-devel
----- Original Message -----
----- Original Message -----
From: "Ayal Baron" <abaron@redhat.com> To: "Itamar Heim" <iheim@redhat.com> Cc: engine-devel@ovirt.org, vdsm-devel@lists.fedorahosted.org Sent: Sunday, March 17, 2013 3:13:09 PM Subject: Re: [Engine-devel] [vdsm] Proposal VDSM <=> Engine Data Statistics Retrieval Optimization
----- Original Message -----
On 03/13/2013 11:55 PM, Ayal Baron wrote: ...
> The only reason we have this problem is because there is > this > thing against making multiple calls. > > Just split it up. > getVmRuntimeStats() - transient things like mem and cpu% > getVmInformation() - (semi)static things like > disk\networking > layout > etc. > Each updated at different intervals.
+1 on splitting the data up into 2 separate API calls. You could potentially add a checksum (md5, or any other way) of the "static" data to getVmRuntimeStats and not bother even with polling the VmInformation if this hasn't changed. Then you could poll as often as you'd like the stats and immediately see if you also need to retrieve VmInfo or not (you rarely would). +1 To Ayal's suggestion except that instead of the engine hashing the data VDSM sends the key which is opaque to the engine. This can be a local timestap or a generation number.
Of course vdsm does the hash, otherwise you'd need to pass all the data to engine which would beat the purpose.
I thought you meant engine will be sending the hash of previous requests per VM to vdsm, then vdsm will reply back with vm's removed, vm's added, and the details for vm's that changed (i.e., engine would be doing something like if-modified-since-checksum per vm). benefit is reducing a round trip. but first would need to split to calls of stats (always changing) and slowly/never changing data.
If vdms accepts the hash then in your method engine would have to periodically call getVmInfo(hash). What I was suggesting is that getVmStats would return vmInfo hash so that we could avoid calling getVmInfo altogether. The stats *always* change so there is no need for checking if that info has changed. What we could do is avoid the split into 2 verbs by calling getVmStats(hash) and then have getVmStats return everything if the hash has changed or only the stats if it hasn't. This would be the least number of roundtrips and avoid the split. If you don't pass a hash it would return everything so this way it's also fully backward compatible.
Actually, I assume we can pass hash 0 (to have vdsm return "everything"). I assume that the chances for md5 on "real data" (i.e - real data that is known to engine) to be 0 are very slim.
We'd need to support hash=None to keep backward compatibility, plus there are no assumptions this way on hash algorithm so why bother with hash=0?
But, we might want to consider that when we add events polling becomes (much) less frequent so maybe it'll be an overkill.
You'd still need to compare versions of the data in vdsm and send only if it changed. If you don't persist what was received last then potentially you could have a monday morning effect where upon on system startup you'd be sending everything. So I still think you'd want to have the hash.
> > ----- Original Message ----- >> From: "Vinzenz Feenstra" <vfeenstr@redhat.com> >> To: vdsm-devel@lists.fedorahosted.org, >> engine-devel@ovirt.org >> Sent: Thursday, March 7, 2013 6:25:54 AM >> Subject: [Engine-devel] Proposal VDSM <=> Engine Data >> Statistics >> Retrieval Optimization >> >> >> Please find the prettier version on the wiki: >> http://www.ovirt.org/Proposal_VDSM_-_Engine_Data_Statistics_Retrieval >> >> Proposal VDSM - Engine Data Statistics Retrieval >> VDSM <=> Engine data retrieval optimization >> Motivation: >> >> >> Currently the RHEVM engine is polling the a lot of data >> from >> VDSM >> every 15 seconds. This should be optimized and the amount >> of >> data >> requested should be more specific. >> >> For each VM the data currently contains much more >> information >> than >> actually needed which blows up the size of the XML content >> quite >> big. We could optimize this by splitting the reply on the >> getVmStats >> based on the request of the engine into sections. For this >> reason >> Omer Frenkel and me have split up the data into parts based >> on >> their >> usage. >> >> This data can and usually does change during the lifetime >> of >> the >> VM. >> Rarely Changed: >> >> >> This data is change not very frequent and it should be >> enough >> to >> update this only once in a while. Most commonly this data >> changes >> after changes made in the UI or after a migration of the VM >> to >> another Host. Status = Running acpiEnable = true vmType = >> kvm >> guestName = W864GUESTAGENTT displayType = qxl guestOs = Win >> 8 >> kvmEnable = true # this should be constant and never >> changed >> pauseCode = NOERR monitorResponse = 0 session = Locked # >> unused >> netIfaces = [{'name': 'Realtek RTL8139C+ Fast Ethernet >> NIC', >> 'inet6': ['fe80::490c:92bb:bbcc:9f87'], 'inet': >> ['10.34.60.148'], >> 'hw': '00:1a:4a:22:3c:db'}] appsList = ['RHEV-Tools 3.2.4', >> 'RHEV-Agent64 3.2.3', 'RHEV-Serial64 3.2.3', >> 'RHEV-Network64 >> 3.2.2', >> 'RHEV-Network64 3.2.3', 'RHEV-Block64 3.2.3', >> 'RHEV-Balloon64 >> 3.2.3', 'RHEV-Balloon64 3.2.2', 'RHEV-Agent64 3.2.2', >> 'RHEV-USB >> 3.2.3', 'RHEV-Block64 3.2.2', 'RHEV-Serial64 3.2.2'] pid = >> 11314 >> guestIPs = 10.34.60.148 # duplicated info displayIp = 0 >> displayPort >> = 5902 displaySecurePort = 5903 username = >> user@W864GUESTAGENTT >> clientIp = lastLogin = 1361976900.67 Often Changed: >> >> >> This data is changed quite often however it is not >> necessary >> to >> update this data every 15 seconds. As this is cumulative >> data >> and >> reflects the current status, and it does not need to be >> snapshotted >> every 15 seconds to retrieve statistics. The data can be >> retrieved >> in much more generous time slices. (e.g. Every 5 minutes) >> network >> = >> {'vnet1': {'macAddr': '00:1a:4a:22:3c:db', 'rxDropped': >> '0', >> 'txDropped': '0', 'rxErrors': '0', 'txRate': '0.0', >> 'rxRate': >> '0.0', >> 'txErrors': '0', 'state': 'unknown', 'speed': '100', >> 'name': >> 'vnet1'}} disksUsage = [{'path': 'c:\\', 'total': >> '64055406592', >> 'fs': 'NTFS', 'used': '19223846912'}, {'path': 'd:\\', >> 'total': >> '3490912256', 'fs': 'UDF', 'used': '3490912256'}] >> timeOffset >> = >> 14422 >> elapsedTime = 68591 hash = 2335461227228498964 statsAge = >> 0.09 >> # >> unused Often Changed but unused >> >> >> This data does not seem to be used in the engine at all. It >> is >> not >> even used in the data warehouse. memoryStats = {'swap_out': >> '0', >> 'majflt': '0', 'mem_free': '1466884', 'swap_in': '0', >> 'pageflt': >> '0', 'mem_total': '2096736', 'mem_unused': '1466884'} >> balloonInfo >> = >> {'balloon_max': 2097152, 'balloon_cur': 2097152} disks = >> {'vda': >> {'readLatency': '0', 'apparentsize': '64424509440', >> 'writeLatency': >> '1754496', 'imageID': >> '28abb923-7b89-4638-84f8-1700f0b76482', >> 'flushLatency': '156549', 'readRate': '0.00', 'truesize': >> '18855059456', 'writeRate': '952.05'}, 'hdc': >> {'readLatency': >> '0', >> 'apparentsize': '0', 'writeLatency': '0', 'flushLatency': >> '0', >> 'readRate': '0.00', 'truesize': '0', 'writeRate': '0.00'}} >> Very >> frequent uppdates needed by webadmin portal: >> >> >> This data is mostly needed for the webadmin portal and >> might >> be >> required to be updated quite often. An exception here is >> the >> statsAge field, which seems to be unused by the Engine. >> This >> data >> could be requested every 15 seconds to keep things as they >> are >> now. >> cpuSys = 2.32 cpuUser = 1.34 memUsage = 30 Proposed >> Solution >> for >> VDSM & Engine: >> >> >> We will introduce new optional parameters to getVmStats, >> getAllVmStats and list to allow a finer grained >> specification >> of >> data which should be included. >> >> Parameter: statsType = <string> (getVmStats, getAllVmStats >> only) >> Allowed values: >> >> * full (default to keep backwards compatibility) >> * app-list (Just send the application list) >> * rare (include everything from rarely changed to very >> frequent) >> * often (include everything from often changed to very >> frequent) >> * frequent (only send the very frequently changed >> items) >> >> >> >> Parameter: clientId = <string> The client id is specified >> by >> the >> client and should be unique however constantly used. >> >> Parameter: diff = <boolean> In combination with the >> clientId >> VDSM >> will send only differences to the previous request from the >> named >> clientId. (if diff=true) >> >> >> Additional Change: >> >> >> Besides the introduction of the new parameters for list, >> getVmStats >> and getAllVmStats it might make sense to include a hash for >> the >> appList into the rarely changed section of the response >> which >> would >> allow to identify changes and avoid having to sent the >> complete >> appList every so often and only if the hash known to the >> client >> is >> outdated. >> >> Note: The appList (Application List) reported by the guest >> agent >> could be fully implemented on request only, as long as the >> guest >> agent installed supports this. As there seems to be a >> request >> to >> have the complete list of installed applications on all >> guests >> this >> data could be quite extensive and a huge list. On the other >> hand >> this data is only rarely visible and therefore it should >> not >> be >> requested all the time and only on demand. Improvement of >> the >> Guest >> Agent: >> >> >> As part of the proposed solution it is necessary to improve >> the >> guest >> agent as well. For the full application list there should >> be >> implemented a caching system which will be fully reactive >> and >> should >> not poll the application list for example all the time. The >> guest >> can create a prepared data file containing all data in the >> JSON >> format (as used for the communication with VDSM via VIO) >> and >> just >> have to read that file from disk and directly sends it to >> VDSM. >> However it is quite possible that this list is to big and >> it >> might >> have to be chunked into pieces. (Multiple messages, which >> would >> have >> to be supported by VDSM then as well) The solution for this >> is >> to >> make VDSM request this data and it will retrieve the data >> necessary >> on request only. -- >> Regards, >> >> Vinzenz Feenstra | Senior Software Engineer >> RedHat Engineering Virtualization R & D >> Phone: +420 532 294 625 >> IRC: vfeenstr or evilissimo >> >> Better technology. Faster innovation. Powered by community >> collaboration. >> See how it works at redhat.com >> _______________________________________________ >> Engine-devel mailing list >> Engine-devel@ovirt.org >> http://lists.ovirt.org/mailman/listinfo/engine-devel >> > _______________________________________________ > vdsm-devel mailing list > vdsm-devel@lists.fedorahosted.org > https://lists.fedorahosted.org/mailman/listinfo/vdsm-devel >
_______________________________________________ vdsm-devel mailing list vdsm-devel@lists.fedorahosted.org https://lists.fedorahosted.org/mailman/listinfo/vdsm-devel
_______________________________________________ Engine-devel mailing list Engine-devel@ovirt.org http://lists.ovirt.org/mailman/listinfo/engine-devel
This is a multi-part message in MIME format. --------------070408040808030806010106 Content-Type: text/plain; charset=UTF-8; format=flowed Content-Transfer-Encoding: 7bit I have left this for a while without continuing because I had to focus on other things. However this is still in progress :-) On 03/13/2013 10:55 PM, Ayal Baron wrote:
----- Original Message -----
----- Original Message -----
From: "Ayal Baron" <abaron@redhat.com> To: "Saggi Mizrahi" <smizrahi@redhat.com> Cc: engine-devel@ovirt.org, vdsm-devel@lists.fedorahosted.org, "Vinzenz Feenstra" <vfeenstr@redhat.com> Sent: Wednesday, March 13, 2013 5:39:24 PM Subject: Re: [vdsm] [Engine-devel] Proposal VDSM <=> Engine Data Statistics Retrieval Optimization
----- Original Message -----
I am completely against this. It make the return value differ according to input which is a big no no when talking about type safe APIs.
The only reason we have this problem is because there is this thing against making multiple calls.
Which is totally contra productive because multiple calls, if properly split up, will actually lead to less data sent for frequent needed data calls. And the others shall be triggered when necessary.
Just split it up. getVmRuntimeStats() - transient things like mem and cpu% getVmInformation() - (semi)static things like disk\networking layout etc. Each updated at different intervals.
+1 on splitting the data up into 2 separate API calls. You could potentially add a checksum (md5, or any other way) of the "static" data to getVmRuntimeStats and not bother even with polling the VmInformation if this hasn't changed. Then you could poll as often as you'd like the stats and immediately see if you also need to retrieve VmInfo or not (you rarely would). +1 To Ayal's suggestion except that instead of the engine hashing the data VDSM sends the key which is opaque to the engine. This can be a local timestap or a generation number. Of course vdsm does the hash, otherwise you'd need to pass all the data to engine which would beat the purpose. We need the hash if we can't have dynamic content. Generation numbers aren't really helpful as every call aggregates the statistics data newly, at the moment at least. But, we might want to consider that when we add events polling becomes (much) less frequent so maybe it'll be an overkill. You'd still need to compare versions of the data in vdsm and send only if it changed. If you don't persist what was received last then potentially you could have a monday morning effect where upon on system startup you'd be sending everything. So I still think you'd want to have the hash. We do a hash already on the XML and include it in the getStats response. Hashes should show enough difference.
Now to the non-dynamic responses and 'type-safe' API: If we would go for non dynamic responses we would need for sure 5 new API calls to achieve some gain on the amount of data sent. *getAllVmRuntimeStats() "returns a map of vmId/data pairs for all vms"* # All the time changing data which is needed by the oVirt Engine, or so often changing that it does not make sense # to place it anywhere else { VmId: { cpuSys --> Could be potentially summarized cpuUser -/ memUsage elapsedTime, status statsAge hashes = { conf, # Hased information of the XML (This one is called "hash" in getStats()) info, # Hashed information of semi static items statusHash: # Hashed information of items with are likely to change however not that often guestDetails: # Hashed value of the guest details (applicationList, network information) } } **getVmStatuses([vmId1, vmId2, ...])*****"Returns a vmId/data pair for each vm requested"** *# This data does not change that often and can be retrieved on demand once the hash changes return { vmId: { timeOffset, monitorResponse clientIp, lastLogin, username, session, guestIPs, } } *getAllVmDeviceStatistics():**"Returns a vmId/data pair for all vms"* # This data has to be requested all the time however in lower intervals (e.g. every 5 minutes) # And is usually needed for all the VMs anyway return { vmId: { network, disksUsage, # Might be improved by summarizing? disks, balloonInfo, memoryStats } } *getVmInfo([vmId1, vmId2, ...]) "Returns a vmId/data pair for each vm requested" * # Basically this should be almost constant, except if there have been changes like migrations, pausing, errors etc return { vmId: { acpiEnable, vmType, guestName, guestOS, kvmEnable, pauseCode, displayIp, displayPort, displaySecurePort, pid, } } *getVmGuestDetails*([vmId1, vmId2, ...]) # Data which changes seldom and these changes can be reflected in the hash when this needs to be requested # This data is really only necessary when it really has been changed or needs to be refreshed for whatever reason. return { vmId: { appsList, netIfaces, } }
----- Original Message -----
From: "Vinzenz Feenstra" <vfeenstr@redhat.com> To: vdsm-devel@lists.fedorahosted.org, engine-devel@ovirt.org Sent: Thursday, March 7, 2013 6:25:54 AM Subject: [Engine-devel] Proposal VDSM <=> Engine Data Statistics Retrieval Optimization
Please find the prettier version on the wiki: http://www.ovirt.org/Proposal_VDSM_-_Engine_Data_Statistics_Retrieval
Proposal VDSM - Engine Data Statistics Retrieval VDSM <=> Engine data retrieval optimization Motivation:
Currently the RHEVM engine is polling the a lot of data from VDSM every 15 seconds. This should be optimized and the amount of data requested should be more specific.
For each VM the data currently contains much more information than actually needed which blows up the size of the XML content quite big. We could optimize this by splitting the reply on the getVmStats based on the request of the engine into sections. For this reason Omer Frenkel and me have split up the data into parts based on their usage.
This data can and usually does change during the lifetime of the VM. Rarely Changed:
This data is change not very frequent and it should be enough to update this only once in a while. Most commonly this data changes after changes made in the UI or after a migration of the VM to another Host. Status = Running acpiEnable = true vmType = kvm guestName = W864GUESTAGENTT displayType = qxl guestOs = Win 8 kvmEnable = true # this should be constant and never changed pauseCode = NOERR monitorResponse = 0 session = Locked # unused netIfaces = [{'name': 'Realtek RTL8139C+ Fast Ethernet NIC', 'inet6': ['fe80::490c:92bb:bbcc:9f87'], 'inet': ['10.34.60.148'], 'hw': '00:1a:4a:22:3c:db'}] appsList = ['RHEV-Tools 3.2.4', 'RHEV-Agent64 3.2.3', 'RHEV-Serial64 3.2.3', 'RHEV-Network64 3.2.2', 'RHEV-Network64 3.2.3', 'RHEV-Block64 3.2.3', 'RHEV-Balloon64 3.2.3', 'RHEV-Balloon64 3.2.2', 'RHEV-Agent64 3.2.2', 'RHEV-USB 3.2.3', 'RHEV-Block64 3.2.2', 'RHEV-Serial64 3.2.2'] pid = 11314 guestIPs = 10.34.60.148 # duplicated info displayIp = 0 displayPort = 5902 displaySecurePort = 5903 username = user@W864GUESTAGENTT clientIp = lastLogin = 1361976900.67 Often Changed:
This data is changed quite often however it is not necessary to update this data every 15 seconds. As this is cumulative data and reflects the current status, and it does not need to be snapshotted every 15 seconds to retrieve statistics. The data can be retrieved in much more generous time slices. (e.g. Every 5 minutes) network = {'vnet1': {'macAddr': '00:1a:4a:22:3c:db', 'rxDropped': '0', 'txDropped': '0', 'rxErrors': '0', 'txRate': '0.0', 'rxRate': '0.0', 'txErrors': '0', 'state': 'unknown', 'speed': '100', 'name': 'vnet1'}} disksUsage = [{'path': 'c:\\', 'total': '64055406592', 'fs': 'NTFS', 'used': '19223846912'}, {'path': 'd:\\', 'total': '3490912256', 'fs': 'UDF', 'used': '3490912256'}] timeOffset = 14422 elapsedTime = 68591 hash = 2335461227228498964 statsAge = 0.09 # unused Often Changed but unused
This data does not seem to be used in the engine at all. It is not even used in the data warehouse. memoryStats = {'swap_out': '0', 'majflt': '0', 'mem_free': '1466884', 'swap_in': '0', 'pageflt': '0', 'mem_total': '2096736', 'mem_unused': '1466884'} balloonInfo = {'balloon_max': 2097152, 'balloon_cur': 2097152} disks = {'vda': {'readLatency': '0', 'apparentsize': '64424509440', 'writeLatency': '1754496', 'imageID': '28abb923-7b89-4638-84f8-1700f0b76482', 'flushLatency': '156549', 'readRate': '0.00', 'truesize': '18855059456', 'writeRate': '952.05'}, 'hdc': {'readLatency': '0', 'apparentsize': '0', 'writeLatency': '0', 'flushLatency': '0', 'readRate': '0.00', 'truesize': '0', 'writeRate': '0.00'}} Very frequent uppdates needed by webadmin portal:
This data is mostly needed for the webadmin portal and might be required to be updated quite often. An exception here is the statsAge field, which seems to be unused by the Engine. This data could be requested every 15 seconds to keep things as they are now. cpuSys = 2.32 cpuUser = 1.34 memUsage = 30 Proposed Solution for VDSM & Engine:
We will introduce new optional parameters to getVmStats, getAllVmStats and list to allow a finer grained specification of data which should be included.
Parameter: statsType = <string> (getVmStats, getAllVmStats only) Allowed values:
* full (default to keep backwards compatibility) * app-list (Just send the application list) * rare (include everything from rarely changed to very frequent) * often (include everything from often changed to very frequent) * frequent (only send the very frequently changed items)
Parameter: clientId = <string> The client id is specified by the client and should be unique however constantly used.
Parameter: diff = <boolean> In combination with the clientId VDSM will send only differences to the previous request from the named clientId. (if diff=true)
Additional Change:
Besides the introduction of the new parameters for list, getVmStats and getAllVmStats it might make sense to include a hash for the appList into the rarely changed section of the response which would allow to identify changes and avoid having to sent the complete appList every so often and only if the hash known to the client is outdated.
Note: The appList (Application List) reported by the guest agent could be fully implemented on request only, as long as the guest agent installed supports this. As there seems to be a request to have the complete list of installed applications on all guests this data could be quite extensive and a huge list. On the other hand this data is only rarely visible and therefore it should not be requested all the time and only on demand. Improvement of the Guest Agent:
As part of the proposed solution it is necessary to improve the guest agent as well. For the full application list there should be implemented a caching system which will be fully reactive and should not poll the application list for example all the time. The guest can create a prepared data file containing all data in the JSON format (as used for the communication with VDSM via VIO) and just have to read that file from disk and directly sends it to VDSM. However it is quite possible that this list is to big and it might have to be chunked into pieces. (Multiple messages, which would have to be supported by VDSM then as well) The solution for this is to make VDSM request this data and it will retrieve the data necessary on request only. -- Regards,
Vinzenz Feenstra | Senior Software Engineer RedHat Engineering Virtualization R & D Phone: +420 532 294 625 IRC: vfeenstr or evilissimo
Better technology. Faster innovation. Powered by community collaboration. See how it works at redhat.com _______________________________________________ Engine-devel mailing list Engine-devel@ovirt.org http://lists.ovirt.org/mailman/listinfo/engine-devel
_______________________________________________ vdsm-devel mailing list vdsm-devel@lists.fedorahosted.org https://lists.fedorahosted.org/mailman/listinfo/vdsm-devel
-- Regards, Vinzenz Feenstra | Senior Software Engineer RedHat Engineering Virtualization R & D Phone: +420 532 294 625 IRC: vfeenstr or evilissimo Better technology. Faster innovation. Powered by community collaboration. See how it works at redhat.com --------------070408040808030806010106 Content-Type: text/html; charset=UTF-8 Content-Transfer-Encoding: 8bit <html> <head> <meta content="text/html; charset=UTF-8" http-equiv="Content-Type"> </head> <body bgcolor="#FFFFFF" text="#000000"> <div class="moz-cite-prefix">I have left this for a while without continuing because I had to focus on other things. However this is still in progress :-)<br> On 03/13/2013 10:55 PM, Ayal Baron wrote:<br> </div> <blockquote cite="mid:25893083.7640290.1363211722007.JavaMail.root@redhat.com" type="cite"> <pre wrap=""> ----- Original Message ----- </pre> <blockquote type="cite"> <pre wrap=""> ----- Original Message ----- </pre> <blockquote type="cite"> <pre wrap="">From: "Ayal Baron" <a class="moz-txt-link-rfc2396E" href="mailto:abaron@redhat.com"><abaron@redhat.com></a> To: "Saggi Mizrahi" <a class="moz-txt-link-rfc2396E" href="mailto:smizrahi@redhat.com"><smizrahi@redhat.com></a> Cc: <a class="moz-txt-link-abbreviated" href="mailto:engine-devel@ovirt.org">engine-devel@ovirt.org</a>, <a class="moz-txt-link-abbreviated" href="mailto:vdsm-devel@lists.fedorahosted.org">vdsm-devel@lists.fedorahosted.org</a>, "Vinzenz Feenstra" <a class="moz-txt-link-rfc2396E" href="mailto:vfeenstr@redhat.com"><vfeenstr@redhat.com></a> Sent: Wednesday, March 13, 2013 5:39:24 PM Subject: Re: [vdsm] [Engine-devel] Proposal VDSM <=> Engine Data Statistics Retrieval Optimization ----- Original Message ----- </pre> <blockquote type="cite"> <pre wrap="">I am completely against this. It make the return value differ according to input which is a big no no when talking about type safe APIs. The only reason we have this problem is because there is this thing against making multiple calls.</pre> </blockquote> </blockquote> </blockquote> </blockquote> Which is totally contra productive because multiple calls, if properly split up, will actually lead to less data sent for frequent needed data calls. And the others shall be triggered when necessary.<br> <blockquote cite="mid:25893083.7640290.1363211722007.JavaMail.root@redhat.com" type="cite"> <blockquote type="cite"> <blockquote type="cite"> <blockquote type="cite"> <pre wrap=""> Just split it up. getVmRuntimeStats() - transient things like mem and cpu% getVmInformation() - (semi)static things like disk\networking layout etc. Each updated at different intervals. </pre> </blockquote> <pre wrap=""> +1 on splitting the data up into 2 separate API calls. You could potentially add a checksum (md5, or any other way) of the "static" data to getVmRuntimeStats and not bother even with polling the VmInformation if this hasn't changed. Then you could poll as often as you'd like the stats and immediately see if you also need to retrieve VmInfo or not (you rarely would). </pre> </blockquote> <pre wrap="">+1 To Ayal's suggestion except that instead of the engine hashing the data VDSM sends the key which is opaque to the engine. This can be a local timestap or a generation number. </pre> </blockquote> <pre wrap=""> Of course vdsm does the hash, otherwise you'd need to pass all the data to engine which would beat the purpose.</pre> </blockquote> We need the hash if we can't have dynamic content. Generation numbers aren't really helpful as every call aggregates the statistics data newly, at the moment at least. <blockquote cite="mid:25893083.7640290.1363211722007.JavaMail.root@redhat.com" type="cite"> <blockquote type="cite"> <pre wrap=""> But, we might want to consider that when we add events polling becomes (much) less frequent so maybe it'll be an overkill. </pre> </blockquote> <pre wrap=""> You'd still need to compare versions of the data in vdsm and send only if it changed. If you don't persist what was received last then potentially you could have a monday morning effect where upon on system startup you'd be sending everything. So I still think you'd want to have the hash.</pre> </blockquote> We do a hash already on the XML and include it in the getStats response.<br> Hashes should show enough difference.<br> <br> Now to the non-dynamic responses and 'type-safe' API: If we would go for non dynamic responses we would need for sure 5 new API calls to achieve some gain on the amount of data sent.<br> <br> <b>getAllVmRuntimeStats() "returns a map of vmId/data pairs for all vms"</b><br> # All the time changing data which is needed by the oVirt Engine, or so often changing that it does not make sense<br> # to place it anywhere else<br> {<br> VmId: {<br> cpuSys --> Could be potentially summarized<br> cpuUser -/<br> memUsage<br> elapsedTime,<br> status<br> statsAge<br> <br> hashes = {<br> conf, # Hased information of the XML (This one is called "hash" in getStats())<br> info, # Hashed information of semi static items<br> statusHash: # Hashed information of items with are likely to change however not that often<br> guestDetails: # Hashed value of the guest details (applicationList, network information)<br> }<br> }<br> <br> <b><b>getVmStatuses([vmId1, vmId2, ...])</b></b><b><b><b> "Returns a vmId/data pair for each vm requested"</b></b><br> </b># This data does not change that often and can be retrieved on demand once the hash changes<br> return {<br> vmId: {<br> timeOffset,<br> monitorResponse<br> clientIp,<br> lastLogin,<br> username,<br> session,<br> guestIPs,<br> }<br> }<br> <br> <b>getAllVmDeviceStatistics():</b><b> "Returns a vmId/data pair for all vms"</b><br> # This data has to be requested all the time however in lower intervals (e.g. every 5 minutes)<br> # And is usually needed for all the VMs anyway<br> return {<br> vmId: {<br> network,<br> disksUsage, # Might be improved by summarizing?<br> disks,<br> balloonInfo,<br> memoryStats<br> }<br> }<br> <br> <b>getVmInfo([vmId1, vmId2, ...]) "Returns a vmId/data pair for each vm requested"<br> </b> # Basically this should be almost constant, except if there have been changes like migrations, pausing, errors etc<br> return {<br> vmId: {<br> acpiEnable,<br> vmType,<br> guestName,<br> guestOS,<br> kvmEnable,<br> pauseCode,<br> displayIp,<br> displayPort,<br> displaySecurePort,<br> pid,<br> }<br> }<br> <br> <b>getVmGuestDetails</b>([vmId1, vmId2, ...])<br> # Data which changes seldom and these changes can be reflected in the hash when this needs to be requested<br> # This data is really only necessary when it really has been changed or needs to be refreshed for whatever reason. <br> return {<br> vmId: {<br> appsList,<br> netIfaces,<br> }<br> }<br> <br> <blockquote cite="mid:25893083.7640290.1363211722007.JavaMail.root@redhat.com" type="cite"> <pre wrap=""> </pre> <blockquote type="cite"> <pre wrap=""> </pre> <blockquote type="cite"> <pre wrap=""> </pre> <blockquote type="cite"> <pre wrap=""> ----- Original Message ----- </pre> <blockquote type="cite"> <pre wrap="">From: "Vinzenz Feenstra" <a class="moz-txt-link-rfc2396E" href="mailto:vfeenstr@redhat.com"><vfeenstr@redhat.com></a> To: <a class="moz-txt-link-abbreviated" href="mailto:vdsm-devel@lists.fedorahosted.org">vdsm-devel@lists.fedorahosted.org</a>, <a class="moz-txt-link-abbreviated" href="mailto:engine-devel@ovirt.org">engine-devel@ovirt.org</a> Sent: Thursday, March 7, 2013 6:25:54 AM Subject: [Engine-devel] Proposal VDSM <=> Engine Data Statistics Retrieval Optimization Please find the prettier version on the wiki: <a class="moz-txt-link-freetext" href="http://www.ovirt.org/Proposal_VDSM_-_Engine_Data_Statistics_Retrieval">http://www.ovirt.org/Proposal_VDSM_-_Engine_Data_Statistics_Retrieval</a> Proposal VDSM - Engine Data Statistics Retrieval VDSM <=> Engine data retrieval optimization Motivation: Currently the RHEVM engine is polling the a lot of data from VDSM every 15 seconds. This should be optimized and the amount of data requested should be more specific. For each VM the data currently contains much more information than actually needed which blows up the size of the XML content quite big. We could optimize this by splitting the reply on the getVmStats based on the request of the engine into sections. For this reason Omer Frenkel and me have split up the data into parts based on their usage. This data can and usually does change during the lifetime of the VM. Rarely Changed: This data is change not very frequent and it should be enough to update this only once in a while. Most commonly this data changes after changes made in the UI or after a migration of the VM to another Host. Status = Running acpiEnable = true vmType = kvm guestName = W864GUESTAGENTT displayType = qxl guestOs = Win 8 kvmEnable = true # this should be constant and never changed pauseCode = NOERR monitorResponse = 0 session = Locked # unused netIfaces = [{'name': 'Realtek RTL8139C+ Fast Ethernet NIC', 'inet6': ['fe80::490c:92bb:bbcc:9f87'], 'inet': ['10.34.60.148'], 'hw': '00:1a:4a:22:3c:db'}] appsList = ['RHEV-Tools 3.2.4', 'RHEV-Agent64 3.2.3', 'RHEV-Serial64 3.2.3', 'RHEV-Network64 3.2.2', 'RHEV-Network64 3.2.3', 'RHEV-Block64 3.2.3', 'RHEV-Balloon64 3.2.3', 'RHEV-Balloon64 3.2.2', 'RHEV-Agent64 3.2.2', 'RHEV-USB 3.2.3', 'RHEV-Block64 3.2.2', 'RHEV-Serial64 3.2.2'] pid = 11314 guestIPs = 10.34.60.148 # duplicated info displayIp = 0 displayPort = 5902 displaySecurePort = 5903 username = user@W864GUESTAGENTT clientIp = lastLogin = 1361976900.67 Often Changed: This data is changed quite often however it is not necessary to update this data every 15 seconds. As this is cumulative data and reflects the current status, and it does not need to be snapshotted every 15 seconds to retrieve statistics. The data can be retrieved in much more generous time slices. (e.g. Every 5 minutes) network = {'vnet1': {'macAddr': '00:1a:4a:22:3c:db', 'rxDropped': '0', 'txDropped': '0', 'rxErrors': '0', 'txRate': '0.0', 'rxRate': '0.0', 'txErrors': '0', 'state': 'unknown', 'speed': '100', 'name': 'vnet1'}} disksUsage = [{'path': 'c:\\', 'total': '64055406592', 'fs': 'NTFS', 'used': '19223846912'}, {'path': 'd:\\', 'total': '3490912256', 'fs': 'UDF', 'used': '3490912256'}] timeOffset = 14422 elapsedTime = 68591 hash = 2335461227228498964 statsAge = 0.09 # unused Often Changed but unused This data does not seem to be used in the engine at all. It is not even used in the data warehouse. memoryStats = {'swap_out': '0', 'majflt': '0', 'mem_free': '1466884', 'swap_in': '0', 'pageflt': '0', 'mem_total': '2096736', 'mem_unused': '1466884'} balloonInfo = {'balloon_max': 2097152, 'balloon_cur': 2097152} disks = {'vda': {'readLatency': '0', 'apparentsize': '64424509440', 'writeLatency': '1754496', 'imageID': '28abb923-7b89-4638-84f8-1700f0b76482', 'flushLatency': '156549', 'readRate': '0.00', 'truesize': '18855059456', 'writeRate': '952.05'}, 'hdc': {'readLatency': '0', 'apparentsize': '0', 'writeLatency': '0', 'flushLatency': '0', 'readRate': '0.00', 'truesize': '0', 'writeRate': '0.00'}} Very frequent uppdates needed by webadmin portal: This data is mostly needed for the webadmin portal and might be required to be updated quite often. An exception here is the statsAge field, which seems to be unused by the Engine. This data could be requested every 15 seconds to keep things as they are now. cpuSys = 2.32 cpuUser = 1.34 memUsage = 30 Proposed Solution for VDSM & Engine: We will introduce new optional parameters to getVmStats, getAllVmStats and list to allow a finer grained specification of data which should be included. Parameter: statsType = <string> (getVmStats, getAllVmStats only) Allowed values: * full (default to keep backwards compatibility) * app-list (Just send the application list) * rare (include everything from rarely changed to very frequent) * often (include everything from often changed to very frequent) * frequent (only send the very frequently changed items) Parameter: clientId = <string> The client id is specified by the client and should be unique however constantly used. Parameter: diff = <boolean> In combination with the clientId VDSM will send only differences to the previous request from the named clientId. (if diff=true) Additional Change: Besides the introduction of the new parameters for list, getVmStats and getAllVmStats it might make sense to include a hash for the appList into the rarely changed section of the response which would allow to identify changes and avoid having to sent the complete appList every so often and only if the hash known to the client is outdated. Note: The appList (Application List) reported by the guest agent could be fully implemented on request only, as long as the guest agent installed supports this. As there seems to be a request to have the complete list of installed applications on all guests this data could be quite extensive and a huge list. On the other hand this data is only rarely visible and therefore it should not be requested all the time and only on demand. Improvement of the Guest Agent: As part of the proposed solution it is necessary to improve the guest agent as well. For the full application list there should be implemented a caching system which will be fully reactive and should not poll the application list for example all the time. The guest can create a prepared data file containing all data in the JSON format (as used for the communication with VDSM via VIO) and just have to read that file from disk and directly sends it to VDSM. However it is quite possible that this list is to big and it might have to be chunked into pieces. (Multiple messages, which would have to be supported by VDSM then as well) The solution for this is to make VDSM request this data and it will retrieve the data necessary on request only. -- Regards, Vinzenz Feenstra | Senior Software Engineer RedHat Engineering Virtualization R & D Phone: +420 532 294 625 IRC: vfeenstr or evilissimo Better technology. Faster innovation. Powered by community collaboration. See how it works at redhat.com _______________________________________________ Engine-devel mailing list <a class="moz-txt-link-abbreviated" href="mailto:Engine-devel@ovirt.org">Engine-devel@ovirt.org</a> <a class="moz-txt-link-freetext" href="http://lists.ovirt.org/mailman/listinfo/engine-devel">http://lists.ovirt.org/mailman/listinfo/engine-devel</a> </pre> </blockquote> <pre wrap="">_______________________________________________ vdsm-devel mailing list <a class="moz-txt-link-abbreviated" href="mailto:vdsm-devel@lists.fedorahosted.org">vdsm-devel@lists.fedorahosted.org</a> <a class="moz-txt-link-freetext" href="https://lists.fedorahosted.org/mailman/listinfo/vdsm-devel">https://lists.fedorahosted.org/mailman/listinfo/vdsm-devel</a> </pre> </blockquote> <pre wrap=""> </pre> </blockquote> <pre wrap=""> </pre> </blockquote> </blockquote> <br> <br> <pre class="moz-signature" cols="72">-- Regards, Vinzenz Feenstra | Senior Software Engineer RedHat Engineering Virtualization R & D Phone: +420 532 294 625 IRC: vfeenstr or evilissimo Better technology. Faster innovation. Powered by community collaboration. See how it works at redhat.com</pre> </body> </html> --------------070408040808030806010106--
----- Original Message -----
I have left this for a while without continuing because I had to focus on other things. However this is still in progress :-)
Are you writing patches? (if so, what solution are you pursuing)
On 03/13/2013 10:55 PM, Ayal Baron wrote:
----- Original Message -----
----- Original Message -----
From: "Ayal Baron" <abaron@redhat.com> To: "Saggi Mizrahi" <smizrahi@redhat.com> Cc: engine-devel@ovirt.org, vdsm-devel@lists.fedorahosted.org, "Vinzenz Feenstra" <vfeenstr@redhat.com> Sent: Wednesday, March 13, 2013 5:39:24 PM Subject: Re: [vdsm] [Engine-devel] Proposal VDSM <=> Engine Data Statistics Retrieval Optimization
----- Original Message -----
I am completely against this. It make the return value differ according to input which is a big no no when talking about type safe APIs.
The only reason we have this problem is because there is this thing against making multiple calls.
Which is totally contra productive because multiple calls, if properly split up, will actually lead to less data sent for frequent needed data calls. And the others shall be triggered when necessary.
Just split it up. getVmRuntimeStats() - transient things like mem and cpu% getVmInformation() - (semi)static things like disk\networking layout etc. Each updated at different intervals.
+1 on splitting the data up into 2 separate API calls. You could potentially add a checksum (md5, or any other way) of the "static" data to getVmRuntimeStats and not bother even with polling the VmInformation if this hasn't changed. Then you could poll as often as you'd like the stats and immediately see if you also need to retrieve VmInfo or not (you rarely would). +1 To Ayal's suggestion except that instead of the engine hashing the data VDSM sends the key which is opaque to the engine. This can be a local timestap or a generation number. Of course vdsm does the hash, otherwise you'd need to pass all the data to engine which would beat the purpose. We need the hash if we can't have dynamic content. Generation numbers aren't really helpful as every call aggregates the statistics data newly, at the moment at least. But, we might want to consider that when we add events polling becomes (much) less frequent so maybe it'll be an overkill. You'd still need to compare versions of the data in vdsm and send only if it changed. If you don't persist what was received last then potentially you could have a monday morning effect where upon on system startup you'd be sending everything. So I still think you'd want to have the hash. We do a hash already on the XML and include it in the getStats response. Hashes should show enough difference.
Now to the non-dynamic responses and 'type-safe' API: If we would go for non dynamic responses we would need for sure 5 new API calls to achieve some gain on the amount of data sent.
*getAllVmRuntimeStats() "returns a map of vmId/data pairs for all vms"* # All the time changing data which is needed by the oVirt Engine, or so often changing that it does not make sense # to place it anywhere else { VmId: { cpuSys --> Could be potentially summarized cpuUser -/ memUsage elapsedTime, status statsAge
hashes = { conf, # Hased information of the XML (This one is called "hash" in getStats()) info, # Hashed information of semi static items statusHash: # Hashed information of items with are likely to change however not that often guestDetails: # Hashed value of the guest details (applicationList, network information) } }
**getVmStatuses([vmId1, vmId2, ...])*****"Returns a vmId/data pair for each vm requested"** *# This data does not change that often and can be retrieved on demand once the hash changes return { vmId: { timeOffset, monitorResponse clientIp, lastLogin, username, session, guestIPs, } }
*getAllVmDeviceStatistics():**"Returns a vmId/data pair for all vms"* # This data has to be requested all the time however in lower intervals (e.g. every 5 minutes) # And is usually needed for all the VMs anyway return { vmId: { network, disksUsage, # Might be improved by summarizing? disks, balloonInfo, memoryStats } }
*getVmInfo([vmId1, vmId2, ...]) "Returns a vmId/data pair for each vm requested" * # Basically this should be almost constant, except if there have been changes like migrations, pausing, errors etc return { vmId: { acpiEnable, vmType, guestName, guestOS, kvmEnable, pauseCode, displayIp, displayPort, displaySecurePort, pid, } }
*getVmGuestDetails*([vmId1, vmId2, ...]) # Data which changes seldom and these changes can be reflected in the hash when this needs to be requested # This data is really only necessary when it really has been changed or needs to be refreshed for whatever reason. return { vmId: { appsList, netIfaces, } }
----- Original Message -----
From: "Vinzenz Feenstra" <vfeenstr@redhat.com> To: vdsm-devel@lists.fedorahosted.org, engine-devel@ovirt.org Sent: Thursday, March 7, 2013 6:25:54 AM Subject: [Engine-devel] Proposal VDSM <=> Engine Data Statistics Retrieval Optimization
Please find the prettier version on the wiki: http://www.ovirt.org/Proposal_VDSM_-_Engine_Data_Statistics_Retrieval
Proposal VDSM - Engine Data Statistics Retrieval VDSM <=> Engine data retrieval optimization Motivation:
Currently the RHEVM engine is polling the a lot of data from VDSM every 15 seconds. This should be optimized and the amount of data requested should be more specific.
For each VM the data currently contains much more information than actually needed which blows up the size of the XML content quite big. We could optimize this by splitting the reply on the getVmStats based on the request of the engine into sections. For this reason Omer Frenkel and me have split up the data into parts based on their usage.
This data can and usually does change during the lifetime of the VM. Rarely Changed:
This data is change not very frequent and it should be enough to update this only once in a while. Most commonly this data changes after changes made in the UI or after a migration of the VM to another Host. Status = Running acpiEnable = true vmType = kvm guestName = W864GUESTAGENTT displayType = qxl guestOs = Win 8 kvmEnable = true # this should be constant and never changed pauseCode = NOERR monitorResponse = 0 session = Locked # unused netIfaces = [{'name': 'Realtek RTL8139C+ Fast Ethernet NIC', 'inet6': ['fe80::490c:92bb:bbcc:9f87'], 'inet': ['10.34.60.148'], 'hw': '00:1a:4a:22:3c:db'}] appsList = ['RHEV-Tools 3.2.4', 'RHEV-Agent64 3.2.3', 'RHEV-Serial64 3.2.3', 'RHEV-Network64 3.2.2', 'RHEV-Network64 3.2.3', 'RHEV-Block64 3.2.3', 'RHEV-Balloon64 3.2.3', 'RHEV-Balloon64 3.2.2', 'RHEV-Agent64 3.2.2', 'RHEV-USB 3.2.3', 'RHEV-Block64 3.2.2', 'RHEV-Serial64 3.2.2'] pid = 11314 guestIPs = 10.34.60.148 # duplicated info displayIp = 0 displayPort = 5902 displaySecurePort = 5903 username = user@W864GUESTAGENTT clientIp = lastLogin = 1361976900.67 Often Changed:
This data is changed quite often however it is not necessary to update this data every 15 seconds. As this is cumulative data and reflects the current status, and it does not need to be snapshotted every 15 seconds to retrieve statistics. The data can be retrieved in much more generous time slices. (e.g. Every 5 minutes) network = {'vnet1': {'macAddr': '00:1a:4a:22:3c:db', 'rxDropped': '0', 'txDropped': '0', 'rxErrors': '0', 'txRate': '0.0', 'rxRate': '0.0', 'txErrors': '0', 'state': 'unknown', 'speed': '100', 'name': 'vnet1'}} disksUsage = [{'path': 'c:\\', 'total': '64055406592', 'fs': 'NTFS', 'used': '19223846912'}, {'path': 'd:\\', 'total': '3490912256', 'fs': 'UDF', 'used': '3490912256'}] timeOffset = 14422 elapsedTime = 68591 hash = 2335461227228498964 statsAge = 0.09 # unused Often Changed but unused
This data does not seem to be used in the engine at all. It is not even used in the data warehouse. memoryStats = {'swap_out': '0', 'majflt': '0', 'mem_free': '1466884', 'swap_in': '0', 'pageflt': '0', 'mem_total': '2096736', 'mem_unused': '1466884'} balloonInfo = {'balloon_max': 2097152, 'balloon_cur': 2097152} disks = {'vda': {'readLatency': '0', 'apparentsize': '64424509440', 'writeLatency': '1754496', 'imageID': '28abb923-7b89-4638-84f8-1700f0b76482', 'flushLatency': '156549', 'readRate': '0.00', 'truesize': '18855059456', 'writeRate': '952.05'}, 'hdc': {'readLatency': '0', 'apparentsize': '0', 'writeLatency': '0', 'flushLatency': '0', 'readRate': '0.00', 'truesize': '0', 'writeRate': '0.00'}} Very frequent uppdates needed by webadmin portal:
This data is mostly needed for the webadmin portal and might be required to be updated quite often. An exception here is the statsAge field, which seems to be unused by the Engine. This data could be requested every 15 seconds to keep things as they are now. cpuSys = 2.32 cpuUser = 1.34 memUsage = 30 Proposed Solution for VDSM & Engine:
We will introduce new optional parameters to getVmStats, getAllVmStats and list to allow a finer grained specification of data which should be included.
Parameter: statsType = <string> (getVmStats, getAllVmStats only) Allowed values:
* full (default to keep backwards compatibility) * app-list (Just send the application list) * rare (include everything from rarely changed to very frequent) * often (include everything from often changed to very frequent) * frequent (only send the very frequently changed items)
Parameter: clientId = <string> The client id is specified by the client and should be unique however constantly used.
Parameter: diff = <boolean> In combination with the clientId VDSM will send only differences to the previous request from the named clientId. (if diff=true)
Additional Change:
Besides the introduction of the new parameters for list, getVmStats and getAllVmStats it might make sense to include a hash for the appList into the rarely changed section of the response which would allow to identify changes and avoid having to sent the complete appList every so often and only if the hash known to the client is outdated.
Note: The appList (Application List) reported by the guest agent could be fully implemented on request only, as long as the guest agent installed supports this. As there seems to be a request to have the complete list of installed applications on all guests this data could be quite extensive and a huge list. On the other hand this data is only rarely visible and therefore it should not be requested all the time and only on demand. Improvement of the Guest Agent:
As part of the proposed solution it is necessary to improve the guest agent as well. For the full application list there should be implemented a caching system which will be fully reactive and should not poll the application list for example all the time. The guest can create a prepared data file containing all data in the JSON format (as used for the communication with VDSM via VIO) and just have to read that file from disk and directly sends it to VDSM. However it is quite possible that this list is to big and it might have to be chunked into pieces. (Multiple messages, which would have to be supported by VDSM then as well) The solution for this is to make VDSM request this data and it will retrieve the data necessary on request only. -- Regards,
Vinzenz Feenstra | Senior Software Engineer RedHat Engineering Virtualization R & D Phone: +420 532 294 625 IRC: vfeenstr or evilissimo
Better technology. Faster innovation. Powered by community collaboration. See how it works at redhat.com _______________________________________________ Engine-devel mailing list Engine-devel@ovirt.org http://lists.ovirt.org/mailman/listinfo/engine-devel
_______________________________________________ vdsm-devel mailing list vdsm-devel@lists.fedorahosted.org https://lists.fedorahosted.org/mailman/listinfo/vdsm-devel
-- Regards,
Vinzenz Feenstra | Senior Software Engineer RedHat Engineering Virtualization R & D Phone: +420 532 294 625 IRC: vfeenstr or evilissimo
Better technology. Faster innovation. Powered by community collaboration. See how it works at redhat.com
On 04/22/2013 10:39 PM, Ayal Baron wrote:
----- Original Message -----
I have left this for a while without continuing because I had to focus on other things. However this is still in progress :-) Are you writing patches? (if so, what solution are you pursuing)
I haven't started writing patches for this particular thing yet, however I want to, but before I want to know that the approach will be acceptable and so I won't have to throw away work because it's not accepted by the project that way. My ultimate goal is to reduce the overall traffic and the amount of data the engine has to parse on a per Host basis. This can be achieved without having push notifications working already. Imagine that oVirt engine currently has to parse for 1000 VMs running, ~16MiB of XML data and that every 15 seconds. And that's just the result from getAllVmStats. Additionally it's polling every 2-3 seconds 'list' to retrieve the VmIds and statuses. Now I have split the values which the engine would return in getVmStats into groups based on how likely the are to change AND their size. getAllVmRuntimeStats() for example, returns the hashes to the different parts where it makes sense, however it won't give you the devices for example. Because the engine still has to poll that every 5 minutes or so to retrieve stats for the the data warehouse. I am personally for having push notifications properly implemented and working however the base of this idea is reducing parser and validation work and will help the engine backend to scale better. Especially it should theoretically reduce the load on the database as well, at least theoretically (I have no numbers for this one ;-))
On 03/13/2013 10:55 PM, Ayal Baron wrote:
----- Original Message -----
----- Original Message -----
From: "Ayal Baron" <abaron@redhat.com> To: "Saggi Mizrahi" <smizrahi@redhat.com> Cc: engine-devel@ovirt.org, vdsm-devel@lists.fedorahosted.org, "Vinzenz Feenstra" <vfeenstr@redhat.com> Sent: Wednesday, March 13, 2013 5:39:24 PM Subject: Re: [vdsm] [Engine-devel] Proposal VDSM <=> Engine Data Statistics Retrieval Optimization
I am completely against this. It make the return value differ according to input which is a big no no when talking about type safe APIs.
The only reason we have this problem is because there is this thing against making multiple calls. Which is totally contra productive because multiple calls, if properly split up, will actually lead to less data sent for frequent needed data calls. And the others shall be triggered when necessary. Just split it up. getVmRuntimeStats() - transient things like mem and cpu% getVmInformation() - (semi)static things like disk\networking layout etc. Each updated at different intervals. +1 on splitting the data up into 2 separate API calls. You could potentially add a checksum (md5, or any other way) of the "static" data to getVmRuntimeStats and not bother even with polling
----- Original Message ----- the VmInformation if this hasn't changed. Then you could poll as often as you'd like the stats and immediately see if you also need to retrieve VmInfo or not (you rarely would). +1 To Ayal's suggestion except that instead of the engine hashing the data VDSM sends the key which is opaque to the engine. This can be a local timestap or a generation number. Of course vdsm does the hash, otherwise you'd need to pass all the data to engine which would beat the purpose. We need the hash if we can't have dynamic content. Generation numbers aren't really helpful as every call aggregates the statistics data newly, at the moment at least. But, we might want to consider that when we add events polling becomes (much) less frequent so maybe it'll be an overkill. You'd still need to compare versions of the data in vdsm and send only if it changed. If you don't persist what was received last then potentially you could have a monday morning effect where upon on system startup you'd be sending everything. So I still think you'd want to have the hash. We do a hash already on the XML and include it in the getStats response. Hashes should show enough difference.
Now to the non-dynamic responses and 'type-safe' API: If we would go for non dynamic responses we would need for sure 5 new API calls to achieve some gain on the amount of data sent.
*getAllVmRuntimeStats() "returns a map of vmId/data pairs for all vms"* # All the time changing data which is needed by the oVirt Engine, or so often changing that it does not make sense # to place it anywhere else { VmId: { cpuSys --> Could be potentially summarized cpuUser -/ memUsage elapsedTime, status statsAge
hashes = { conf, # Hased information of the XML (This one is called "hash" in getStats()) info, # Hashed information of semi static items statusHash: # Hashed information of items with are likely to change however not that often guestDetails: # Hashed value of the guest details (applicationList, network information) } }
**getVmStatuses([vmId1, vmId2, ...])*****"Returns a vmId/data pair for each vm requested"** *# This data does not change that often and can be retrieved on demand once the hash changes return { vmId: { timeOffset, monitorResponse clientIp, lastLogin, username, session, guestIPs, } }
*getAllVmDeviceStatistics():**"Returns a vmId/data pair for all vms"* # This data has to be requested all the time however in lower intervals (e.g. every 5 minutes) # And is usually needed for all the VMs anyway return { vmId: { network, disksUsage, # Might be improved by summarizing? disks, balloonInfo, memoryStats } }
*getVmInfo([vmId1, vmId2, ...]) "Returns a vmId/data pair for each vm requested" * # Basically this should be almost constant, except if there have been changes like migrations, pausing, errors etc return { vmId: { acpiEnable, vmType, guestName, guestOS, kvmEnable, pauseCode, displayIp, displayPort, displaySecurePort, pid, } }
*getVmGuestDetails*([vmId1, vmId2, ...]) # Data which changes seldom and these changes can be reflected in the hash when this needs to be requested # This data is really only necessary when it really has been changed or needs to be refreshed for whatever reason. return { vmId: { appsList, netIfaces, } }
----- Original Message ----- > From: "Vinzenz Feenstra" <vfeenstr@redhat.com> > To: vdsm-devel@lists.fedorahosted.org, engine-devel@ovirt.org > Sent: Thursday, March 7, 2013 6:25:54 AM > Subject: [Engine-devel] Proposal VDSM <=> Engine Data > Statistics > Retrieval Optimization > > > Please find the prettier version on the wiki: > http://www.ovirt.org/Proposal_VDSM_-_Engine_Data_Statistics_Retrieval > > Proposal VDSM - Engine Data Statistics Retrieval > VDSM <=> Engine data retrieval optimization > Motivation: > > > Currently the RHEVM engine is polling the a lot of data from > VDSM > every 15 seconds. This should be optimized and the amount of > data > requested should be more specific. > > For each VM the data currently contains much more information > than > actually needed which blows up the size of the XML content > quite > big. We could optimize this by splitting the reply on the > getVmStats > based on the request of the engine into sections. For this > reason > Omer Frenkel and me have split up the data into parts based on > their > usage. > > This data can and usually does change during the lifetime of > the > VM. > Rarely Changed: > > > This data is change not very frequent and it should be enough > to > update this only once in a while. Most commonly this data > changes > after changes made in the UI or after a migration of the VM to > another Host. Status = Running acpiEnable = true vmType = kvm > guestName = W864GUESTAGENTT displayType = qxl guestOs = Win 8 > kvmEnable = true # this should be constant and never changed > pauseCode = NOERR monitorResponse = 0 session = Locked # unused > netIfaces = [{'name': 'Realtek RTL8139C+ Fast Ethernet NIC', > 'inet6': ['fe80::490c:92bb:bbcc:9f87'], 'inet': > ['10.34.60.148'], > 'hw': '00:1a:4a:22:3c:db'}] appsList = ['RHEV-Tools 3.2.4', > 'RHEV-Agent64 3.2.3', 'RHEV-Serial64 3.2.3', 'RHEV-Network64 > 3.2.2', > 'RHEV-Network64 3.2.3', 'RHEV-Block64 3.2.3', 'RHEV-Balloon64 > 3.2.3', 'RHEV-Balloon64 3.2.2', 'RHEV-Agent64 3.2.2', 'RHEV-USB > 3.2.3', 'RHEV-Block64 3.2.2', 'RHEV-Serial64 3.2.2'] pid = > 11314 > guestIPs = 10.34.60.148 # duplicated info displayIp = 0 > displayPort > = 5902 displaySecurePort = 5903 username = user@W864GUESTAGENTT > clientIp = lastLogin = 1361976900.67 Often Changed: > > > This data is changed quite often however it is not necessary to > update this data every 15 seconds. As this is cumulative data > and > reflects the current status, and it does not need to be > snapshotted > every 15 seconds to retrieve statistics. The data can be > retrieved > in much more generous time slices. (e.g. Every 5 minutes) > network > = > {'vnet1': {'macAddr': '00:1a:4a:22:3c:db', 'rxDropped': '0', > 'txDropped': '0', 'rxErrors': '0', 'txRate': '0.0', 'rxRate': > '0.0', > 'txErrors': '0', 'state': 'unknown', 'speed': '100', 'name': > 'vnet1'}} disksUsage = [{'path': 'c:\\', 'total': > '64055406592', > 'fs': 'NTFS', 'used': '19223846912'}, {'path': 'd:\\', 'total': > '3490912256', 'fs': 'UDF', 'used': '3490912256'}] timeOffset = > 14422 > elapsedTime = 68591 hash = 2335461227228498964 statsAge = 0.09 > # > unused Often Changed but unused > > > This data does not seem to be used in the engine at all. It is > not > even used in the data warehouse. memoryStats = {'swap_out': > '0', > 'majflt': '0', 'mem_free': '1466884', 'swap_in': '0', > 'pageflt': > '0', 'mem_total': '2096736', 'mem_unused': '1466884'} > balloonInfo > = > {'balloon_max': 2097152, 'balloon_cur': 2097152} disks = > {'vda': > {'readLatency': '0', 'apparentsize': '64424509440', > 'writeLatency': > '1754496', 'imageID': '28abb923-7b89-4638-84f8-1700f0b76482', > 'flushLatency': '156549', 'readRate': '0.00', 'truesize': > '18855059456', 'writeRate': '952.05'}, 'hdc': {'readLatency': > '0', > 'apparentsize': '0', 'writeLatency': '0', 'flushLatency': '0', > 'readRate': '0.00', 'truesize': '0', 'writeRate': '0.00'}} Very > frequent uppdates needed by webadmin portal: > > > This data is mostly needed for the webadmin portal and might be > required to be updated quite often. An exception here is the > statsAge field, which seems to be unused by the Engine. This > data > could be requested every 15 seconds to keep things as they are > now. > cpuSys = 2.32 cpuUser = 1.34 memUsage = 30 Proposed Solution > for > VDSM & Engine: > > > We will introduce new optional parameters to getVmStats, > getAllVmStats and list to allow a finer grained specification > of > data which should be included. > > Parameter: statsType = <string> (getVmStats, getAllVmStats > only) > Allowed values: > > * full (default to keep backwards compatibility) > * app-list (Just send the application list) > * rare (include everything from rarely changed to very > frequent) > * often (include everything from often changed to very > frequent) > * frequent (only send the very frequently changed items) > > > > Parameter: clientId = <string> The client id is specified by > the > client and should be unique however constantly used. > > Parameter: diff = <boolean> In combination with the clientId > VDSM > will send only differences to the previous request from the > named > clientId. (if diff=true) > > > Additional Change: > > > Besides the introduction of the new parameters for list, > getVmStats > and getAllVmStats it might make sense to include a hash for the > appList into the rarely changed section of the response which > would > allow to identify changes and avoid having to sent the complete > appList every so often and only if the hash known to the client > is > outdated. > > Note: The appList (Application List) reported by the guest > agent > could be fully implemented on request only, as long as the > guest > agent installed supports this. As there seems to be a request > to > have the complete list of installed applications on all guests > this > data could be quite extensive and a huge list. On the other > hand > this data is only rarely visible and therefore it should not be > requested all the time and only on demand. Improvement of the > Guest > Agent: > > > As part of the proposed solution it is necessary to improve the > guest > agent as well. For the full application list there should be > implemented a caching system which will be fully reactive and > should > not poll the application list for example all the time. The > guest > can create a prepared data file containing all data in the JSON > format (as used for the communication with VDSM via VIO) and > just > have to read that file from disk and directly sends it to VDSM. > However it is quite possible that this list is to big and it > might > have to be chunked into pieces. (Multiple messages, which would > have > to be supported by VDSM then as well) The solution for this is > to > make VDSM request this data and it will retrieve the data > necessary > on request only. -- > Regards, > > Vinzenz Feenstra | Senior Software Engineer > RedHat Engineering Virtualization R & D > Phone: +420 532 294 625 > IRC: vfeenstr or evilissimo > > Better technology. Faster innovation. Powered by community > collaboration. > See how it works at redhat.com > _______________________________________________ > Engine-devel mailing list > Engine-devel@ovirt.org > http://lists.ovirt.org/mailman/listinfo/engine-devel > _______________________________________________ vdsm-devel mailing list vdsm-devel@lists.fedorahosted.org https://lists.fedorahosted.org/mailman/listinfo/vdsm-devel
-- Regards,
Vinzenz Feenstra | Senior Software Engineer RedHat Engineering Virtualization R & D Phone: +420 532 294 625 IRC: vfeenstr or evilissimo
Better technology. Faster innovation. Powered by community collaboration. See how it works at redhat.com
-- Regards, Vinzenz Feenstra | Senior Software Engineer RedHat Engineering Virtualization R & D Phone: +420 532 294 625 IRC: vfeenstr or evilissimo Better technology. Faster innovation. Powered by community collaboration. See how it works at redhat.com
On 03/07/2013 01:25 PM, Vinzenz Feenstra wrote:
Often Changed but unused
This data does not seem to be used in the engine at all. It is *not* even used in the data warehouse.
*memoryStats* = {'swap_out': '0', 'majflt': '0', 'mem_free': '1466884', 'swap_in': '0', 'pageflt': '0', 'mem_total': '2096736', 'mem_unused': '1466884'} *balloonInfo* = {'balloon_max': 2097152, 'balloon_cur': 2097152}
these were added for better balloon control iirc. so api needs them for mom probbaly. not sure if engine needs them or not.
*disks* = {'vda': {'readLatency': '0', 'apparentsize': '64424509440', 'writeLatency': '1754496', 'imageID': '28abb923-7b89-4638-84f8-1700f0b76482', 'flushLatency': '156549', 'readRate': '0.00', 'truesize': '18855059456', 'writeRate': '952.05'}, 'hdc': {'readLatency': '0', 'apparentsize': '0', 'writeLatency': '0', 'flushLatency': '0', 'readRate': '0.00', 'truesize': '0', 'writeRate': '0.00'}}
these i remember we added per request to the system - are you sure they are not in the dwh?
participants (8)
-
Ayal Baron -
Dan Kenigsberg -
Itamar Heim -
Mark Wu -
Saggi Mizrahi -
Vinzenz Feenstra -
Yair Zaslavsky -
Yaniv Kaul