https://jira.whamcloud.com This file is an XML representation of an issue en-us 9.4.14 940014 05-12-2023 https://jira.whamcloud.com/browse/LU-5197 Lustre <p>There is a perforamnce regression of "FileRead" metadata operation caused by fixes of <a href="https://jira.whamcloud.com/browse/LU-4398" title="mdt_object_open_lock() may not flush conflicting handles" class="issue-link" data-issue-key="LU-4398"><del>LU-4398</del></a>.<br/> Please see below performance difference with/without <a href="https://jira.whamcloud.com/browse/LU-4398" title="mdt_object_open_lock() may not flush conflicting handles" class="issue-link" data-issue-key="LU-4398"><del>LU-4398</del></a> patches.</p> <p>It's simple mdtest run on single client with single thread.</p> <div class="preformatted panel" style="border-width: 1px;"><div class="preformattedContent panelContent"> <pre>#mpirun -np 1 -ppn 1 -hostfile ./hostfile ./mdtest -n 100000 -i 1 -p 5 -u -v -F -d mdtest.out v2_5_2_RC1 SUMMARY: (of 1 iterations) Operation Max Min Mean Std Dev --------- --- --- ---- ------- File creation : 3478.187 3478.187 3478.187 0.000 File stat : 4514.273 4514.273 4514.273 0.000 File read : 515.113 515.113 515.113 0.000 File removal : 6499.513 6499.513 6499.513 0.000 Tree creation : 3908.951 3908.951 3908.951 0.000 Tree removal : 428.646 428.646 428.646 0.000 v2_5_2_RC1 + revert "LU-4398 mdt: acquire an open lock for write or execute" (97bfe7a3c0fc74fb0e56cbc1ea9cb827fb657b48) SUMMARY: (of 1 iterations) Operation Max Min Mean Std Dev --------- --- --- ---- ------- File creation : 3575.570 3575.570 3575.570 0.000 File stat : 4708.052 4708.052 4708.052 0.000 File read : 5348.987 5348.987 5348.987 0.000 File removal : 6509.834 6509.834 6509.834 0.000 Tree creation : 3816.473 3816.473 3816.473 0.000 Tree removal : 400.143 400.143 400.143 0.000 </pre> </div></div> LU-5197

A performance regression of "FileRead" metadata operation

Bug Critical Resolved Fixed John Hammond Shuichi Ihara HB Sat, 14 Jun 2014 19:32:00 +0000 Tue, 13 Jan 2015 13:55:39 +0000 Mon, 23 Jun 2014 23:17:05 +0000 Lustre 2.6.0 Lustre 2.5.2 Lustre 2.6.0 0 13 <p>John</p> <p>You are the author of the patch in question. Do you have any suggestions here?</p> <p>Thanks</p> <p>Peter</p> <p>Investigating.</p> <p>Ihara,</p> <p>Could you try with <a href="http://review.whamcloud.com/10725?" class="external-link" target="_blank" rel="nofollow noopener">http://review.whamcloud.com/10725?</a></p> <p>Also I see that mdtest_read() calls open() with O_RDWR. Would it be possible to change that to O_RDONLY and rerun?</p> <p>There were a few patches under <a href="https://jira.whamcloud.com/browse/LU-4367" title="unlink performance regression on lustre-2.5.52 client" class="issue-link" data-issue-key="LU-4367"><del>LU-4367</del></a> that also affected the open performance that we were not considering as critical for 2.6.0, but may also help performance here.</p> <p>Have you tested with any of those patches applied, in particular <a href="http://review.whamcloud.com/9696" class="external-link" target="_blank" rel="nofollow noopener">http://review.whamcloud.com/9696</a>. The patch <a href="http://review.whamcloud.com/9697" class="external-link" target="_blank" rel="nofollow noopener">http://review.whamcloud.com/9697</a> also looks very similar to John's patch here.</p> <p>Sorry, dealy of this testing, I got some test resutls today. still bad file read operation.</p> <div class='table-wrap'> <table class='confluenceTable'><tbody> <tr> <td class='confluenceTd'>branch</td> <td class='confluenceTd'> File read (ops/sec)</td> </tr> <tr> <td class='confluenceTd'>master</td> <td class='confluenceTd'> 620 </td> </tr> <tr> <td class='confluenceTd'>master+10725</td> <td class='confluenceTd'> 623</td> </tr> <tr> <td class='confluenceTd'>master+9696</td> <td class='confluenceTd'> 625</td> </tr> </tbody></table> </div> <p>Hi Ihara,</p> <p>Could you provide the results for master + revert of 97bfe7a3c0fc74fb0e56cbc1ea9cb827fb657b48?</p> <p>Thanks,</p> <p>John</p> <p>John,</p> <p>You mean revert 708d85a652a77f85153790e6cca1b7a2b91947cf (Revert "<a href="https://jira.whamcloud.com/browse/LU-4398" title="mdt_object_open_lock() may not flush conflicting handles" class="issue-link" data-issue-key="LU-4398"><del>LU-4398</del></a> mdt: acquire an open lock for write or execute"), right? I couldn't find 97bfe7a3c0fc74fb0e56cbc1ea9cb827fb657b48 in the master.</p> <p>Anyway, here is results of master + revert of it. The FileRead operation is back to normal.</p> <div class="preformatted panel" style="border-width: 1px;"><div class="preformattedContent panelContent"> <pre>SUMMARY: (of 1 iterations) Operation Max Min Mean Std Dev --------- --- --- ---- ------- File creation : 4331.477 4331.477 4331.477 0.000 File stat : 5040.508 5040.508 5040.508 0.000 File read : 5780.921 5780.921 5780.921 0.000 File removal : 7401.880 7401.880 7401.880 0.000 Tree creation : 3184.741 3184.741 3184.741 0.000 Tree removal : 366.283 366.283 366.283 0.000 V-1: Entering timestamp... -- finished at 06/19/2014 23:53:31 -- </pre> </div></div> <p>Patch from <a href="https://jira.whamcloud.com/browse/LU-4398" title="mdt_object_open_lock() may not flush conflicting handles" class="issue-link" data-issue-key="LU-4398"><del>LU-4398</del></a> causing this issue was reverted.</p> Related LU-4367 LU-4398 Development Rank 1|hzwp2v: Rank (Obsolete) 14517 Severity