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-10181 Lustre <p>Various optimization for better DoM performance. So far they are:</p> <ul> <li>glimpse ahead. Issue glimpse from MDT in advance and return size on client along with other attributes.</li> <li>read-on-open. Read file data at OPEN and return to the client</li> <li>lock bits convert instead of cancel. It is separate ticket because it is useful for all files, not just DoM.</li> </ul> LU-10181

DoM performance optimization

Technical task LU-10176 Major Resolved Fixed Mikhail Pershin Mikhail Pershin DoM DoM2 Wed, 1 Nov 2017 15:33:46 +0000 Wed, 15 Jan 2020 06:41:13 +0000 Thu, 9 Aug 2018 19:03:03 +0000 Lustre 2.12.0 0 6 <p>read-on-open patches:<br/> <a href="https://review.whamcloud.com/#/c/23011/" class="external-link" target="_blank" rel="nofollow noopener">https://review.whamcloud.com/#/c/23011/</a></p> <p>I've done quick tests with the patch, on my local VM. There are some numbers:<br/> 1. time grep -R "something" on linux source dir</p> <div class="preformatted panel" style="border-width: 1px;"><div class="preformattedContent panelContent"> <pre>DoM (Read-on-Open) : real 1m6.677s user 0m1.107s sys 0m35.430s real 1m8.711s user 0m1.129s sys 0m35.416s real 1m4.229s user 0m1.071s sys 0m34.549s DoM (no Read-on-Open): real 1m9.945s user 0m0.926s sys 0m35.330s real 1m10.561s user 0m0.930s sys 0m34.905s real 1m11.278s user 0m0.871s sys 0m35.280s OST files (2 stripes): real 1m23.753s user 0m1.033s sys 0m37.412s real 1m27.987s user 0m1.032s sys 0m39.327s real 1m20.697s user 0m0.993s sys 0m35.494s </pre> </div></div> <p>RPC stats shows there are only 593 READ RPCs for about 47697 files with read-on-open, but resulting time is less not much, probably the difference will be seen better on real system, not VM.</p> <p>2. dbench 4</p> <div class="preformatted panel" style="border-width: 1px;"><div class="preformattedContent panelContent"> <pre>DoM (Read-on-Open) : Throughput 42.459 MB/sec 4 clients 4 procs max_latency=1224.798 ms DoM (no Read-on-Open): Throughput 37.2351 MB/sec 4 clients 4 procs max_latency=1163.293 ms OST files (2 stripes): Throughput 28.0458 MB/sec 4 clients 4 procs max_latency=2027.644 ms </pre> </div></div> <p>Dbench shows clear difference in results for DoM in general and Read-on-Open</p> <p> </p> <p>Mike Pershin (mike.pershin@intel.com) uploaded a new patch: <a href="https://review.whamcloud.com/31271" class="external-link" target="_blank" rel="nofollow noopener">https://review.whamcloud.com/31271</a><br/> Subject: <a href="https://jira.whamcloud.com/browse/LU-10181" title="DoM performance optimization" class="issue-link" data-issue-key="LU-10181"><del>LU-10181</del></a> test: debug tests<br/> Project: fs/lustre-release<br/> Branch: master<br/> Current Patch Set: 1<br/> Commit: 30ffbccb384ed29e3b02dc9433213127e5e87dfd</p> <p>Oleg Drokin (oleg.drokin@intel.com) merged in patch <a href="https://review.whamcloud.com/29968/" class="external-link" target="_blank" rel="nofollow noopener">https://review.whamcloud.com/29968/</a><br/> Subject: <a href="https://jira.whamcloud.com/browse/LU-10181" title="DoM performance optimization" class="issue-link" data-issue-key="LU-10181"><del>LU-10181</del></a> mdt: high-priority request handling for DOM<br/> Project: fs/lustre-release<br/> Branch: master<br/> Current Patch Set: <br/> Commit: 9e338ee25609d9bbcc359c37012e6868dd6be5d3</p> <p>Oleg Drokin (oleg.drokin@intel.com) merged in patch <a href="https://review.whamcloud.com/30059/" class="external-link" target="_blank" rel="nofollow noopener">https://review.whamcloud.com/30059/</a><br/> Subject: <a href="https://jira.whamcloud.com/browse/LU-10181" title="DoM performance optimization" class="issue-link" data-issue-key="LU-10181"><del>LU-10181</del></a> tests: add FIO as test for DOM<br/> Project: fs/lustre-release<br/> Branch: master<br/> Current Patch Set: <br/> Commit: 84099c2135e8b11ffd2688a4b4c2529e6e60468f</p> <p>Mike Pershin (mike.pershin@intel.com) uploaded a new patch: <a href="https://review.whamcloud.com/31749" class="external-link" target="_blank" rel="nofollow noopener">https://review.whamcloud.com/31749</a><br/> Subject: <a href="https://jira.whamcloud.com/browse/LU-10181" title="DoM performance optimization" class="issue-link" data-issue-key="LU-10181"><del>LU-10181</del></a> test: add compilebench and DNE tests<br/> Project: fs/lustre-release<br/> Branch: master<br/> Current Patch Set: 1<br/> Commit: 842a88e9adc0679f8f82b8c18af57385639c1ecb</p> <p>Mike Pershin (mike.pershin@intel.com) uploaded a new patch: <a href="https://review.whamcloud.com/31750" class="external-link" target="_blank" rel="nofollow noopener">https://review.whamcloud.com/31750</a><br/> Subject: <a href="https://jira.whamcloud.com/browse/LU-10181" title="DoM performance optimization" class="issue-link" data-issue-key="LU-10181"><del>LU-10181</del></a> mds: init cpt params for mdt IO service<br/> Project: fs/lustre-release<br/> Branch: master<br/> Current Patch Set: 1<br/> Commit: 23497102e83b05c1926b9dff652d6fcb299bfea1</p> <p>Mike Pershin (mike.pershin@intel.com) uploaded a new patch: <a href="https://review.whamcloud.com/31768" class="external-link" target="_blank" rel="nofollow noopener">https://review.whamcloud.com/31768</a><br/> Subject: <a href="https://jira.whamcloud.com/browse/LU-10181" title="DoM performance optimization" class="issue-link" data-issue-key="LU-10181"><del>LU-10181</del></a> mdt: take discard lock at cleanup stage<br/> Project: fs/lustre-release<br/> Branch: master<br/> Current Patch Set: 1<br/> Commit: 283521ae6fb612598552a56271d11d442b7b2e5b</p> <p>Oleg Drokin (green@whamcloud.com) merged in patch <a href="https://review.whamcloud.com/31750/" class="external-link" target="_blank" rel="nofollow noopener">https://review.whamcloud.com/31750/</a><br/> Subject: <a href="https://jira.whamcloud.com/browse/LU-10181" title="DoM performance optimization" class="issue-link" data-issue-key="LU-10181"><del>LU-10181</del></a> mds: init cpt params for mdt IO service<br/> Project: fs/lustre-release<br/> Branch: master<br/> Current Patch Set: <br/> Commit: 3d29ce84a51bcb9593a3d9bca3c570b36b566623</p> <p>Oleg Drokin (green@whamcloud.com) merged in patch <a href="https://review.whamcloud.com/23011/" class="external-link" target="_blank" rel="nofollow noopener">https://review.whamcloud.com/23011/</a><br/> Subject: <a href="https://jira.whamcloud.com/browse/LU-10181" title="DoM performance optimization" class="issue-link" data-issue-key="LU-10181"><del>LU-10181</del></a> mdt: read on open for DoM files<br/> Project: fs/lustre-release<br/> Branch: master<br/> Current Patch Set: <br/> Commit: 13372d6c243ccfc1a0886469f3c0dc048783ed47</p> <p>All current patches are now landed for 2.12</p> Related LU-3285 LU-13136 Development Rank 1|hzzmwn: Rank (Obsolete) 9223372036854775807