<div dir="ltr"><br><div class="gmail_extra"><br><div class="gmail_quote">On Mon, Sep 18, 2017 at 10:50 PM, Martin Polednik <span dir="ltr"><<a href="mailto:mpoledni@redhat.com" target="_blank">mpoledni@redhat.com</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;padding-left:1ex;border-left-color:rgb(204,204,204);border-left-width:1px;border-left-style:solid">The hugepages are no longer a hook, but part of the code base. They</blockquote><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;padding-left:1ex;border-left-color:rgb(204,204,204);border-left-width:1px;border-left-style:solid">
can be configured via engine property `hugepages`, where the value of<br>
property is size of the pages in KiB (1048576 = 1G, 2048 = 2M).<br></blockquote><div><br></div><div><div>Note that the question is about 4.1 and it doesn't seem like this change was backported to the 4.1 branch, right?</div></div><div> </div><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;padding-left:1ex;border-left-color:rgb(204,204,204);border-left-width:1px;border-left-style:solid">
Specifying any other size not supported by the architecture (or just<br>
"1") will use platform's default hugepage size.<br></blockquote><div><br></div><div>I wonder if the end-to-end scenario, i.e., using the engine, of this ever worked.</div><div>This feature is broken on the master branch because of the switch to engine xml and is supposed to be fixed by [1]. However, while testing that fix I noticed that a check done by the scheduler [2] actually prevents us from setting a value that is not supported by the host.</div><div>Is this part really important?</div><div><br></div><div>[1] <a href="https://gerrit.ovirt.org/#/c/81860/">https://gerrit.ovirt.org/#/c/81860/</a></div><div>[2] <a href="https://github.com/oVirt/ovirt-engine/blob/master/backend/manager/modules/bll/src/main/java/org/ovirt/engine/core/bll/scheduling/policyunits/HugePagesFilterPolicyUnit.java#L56">https://github.com/oVirt/ovirt-engine/blob/master/backend/manager/modules/bll/src/main/java/org/ovirt/engine/core/bll/scheduling/policyunits/HugePagesFilterPolicyUnit.java#L56</a></div><div> </div><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;padding-left:1ex;border-left-color:rgb(204,204,204);border-left-width:1px;border-left-style:solid">
<br>
On Mon, Sep 18, 2017 at 11:39 AM, Gianluca Cecchi<br>
<<a href="mailto:gianluca.cecchi@gmail.com">gianluca.cecchi@gmail.com</a>> wrote:<br>
> Hello,<br>
> I would like to go again in deep with what preliminary tested and already<br>
> discussed here:<br>
><br>
> <a href="http://lists.ovirt.org/pipermail/users/2017-April/081320.html" target="_blank" rel="noreferrer">http://lists.ovirt.org/<wbr>pipermail/users/2017-April/<wbr>081320.html</a><br>
><br>
> I'm testing an oVirt node in 4.1.6-pre<br>
><br>
> I don't find the vdsm hook for huge pages; doing a search I get these:<br>
><br>
> vdsm-hook-ethtool-options.<wbr>noarch : Allow setting custom ethtool options for<br>
> vdsm controlled nics<br>
> vdsm-hook-fcoe.noarch : Hook to enable FCoE support<br>
> vdsm-hook-openstacknet.noarch : OpenStack Network vNICs support for VDSM<br>
> vdsm-hook-vfio-mdev.noarch : Hook to enable mdev-capable devices.<br>
> vdsm-hook-vhostmd.noarch : VDSM hook set for interaction with vhostmd<br>
> vdsm-hook-vmfex-dev.noarch : VM-FEX vNIC support for VDSM<br>
><br>
> Did anything change between 4.1.1 and 4.1.5/4.1.6?<br>
><br>
> I'm making preliminary tests with an Oracle RDBMS and HammerDB in both a<br>
> physical server and a "big" VM inside another same hw server configured with<br>
> oVirt.<br>
><br>
> Results are not bad, but I would like to see having huge pages inside the<br>
> guest how could change results.<br>
><br>
> Just for reference:<br>
><br>
> The 2 Physical server are blades with each one:<br>
> 2 sockets, each one with 14 cores and ht enabled, so in total 56<br>
> computational threads<br>
> 256Gb ram<br>
> huge pages enabled<br>
><br>
> VM configured with this virtual hw on one of them:<br>
> 2 sockets, each one with 6 cores and ht so in total 24 computational threads<br>
> 64Gb ram<br>
> no huge pages at the moment<br>
><br>
> Oracle SGA is 32Gb on both physical rdbms and virtual one.<br>
><br>
> Thanks for any insight to test huge pages in guest<br>
><br>
> Gianluca<br>
><br>
> ______________________________<wbr>_________________<br>
> Users mailing list<br>
> <a href="mailto:Users@ovirt.org">Users@ovirt.org</a><br>
> <a href="http://lists.ovirt.org/mailman/listinfo/users" target="_blank" rel="noreferrer">http://lists.ovirt.org/<wbr>mailman/listinfo/users</a><br>
><br>
______________________________<wbr>_________________<br>
Users mailing list<br>
<a href="mailto:Users@ovirt.org">Users@ovirt.org</a><br>
<a href="http://lists.ovirt.org/mailman/listinfo/users" target="_blank" rel="noreferrer">http://lists.ovirt.org/<wbr>mailman/listinfo/users</a><br>
</blockquote></div><br></div></div>