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-9618 Lustre <p>When writing to a region of a file less than file size (either an already existing file or a shared file with multiple writers), writes of less than one page in size must first read in that page.</p> <p>This results in extremely poor performance. For the random I/O case, there's no easy improvements available, but the sequential case can benefit enormously by using readahead to bring in those pages.</p> <p>This can be done by using the normal read code w/readahead from the ll_prepare_partial_page function. This works well in the sequential case (3-10x performance improvement) and has no effect on the random case.</p> <p>Patch &amp; benchmark #s forthcoming shortly.</p> LU-9618

Connect readahead to prep_partial_page to improve small (< 1 page) write performance

Bug Minor Resolved Fixed Patrick Farrell Patrick Farrell patch Wed, 7 Jun 2017 23:18:24 +0000 Fri, 30 Aug 2019 07:30:49 +0000 Sat, 20 Jan 2018 16:31:42 +0000 Lustre 2.11.0 0 5 <p>Patrick Farrell (paf@cray.com) uploaded a new patch: <a href="https://review.whamcloud.com/27544" class="external-link" target="_blank" rel="nofollow noopener">https://review.whamcloud.com/27544</a><br/> Subject: <a href="https://jira.whamcloud.com/browse/LU-9618" title="Connect readahead to prep_partial_page to improve small (&lt; 1 page) write performance" class="issue-link" data-issue-key="LU-9618"><del>LU-9618</del></a> clio: Use readahead for partial page write<br/> Project: fs/lustre-release<br/> Branch: master<br/> Current Patch Set: 1<br/> Commit: 1b9d2d4a523fc7d02245864942246fef7b92e3c5</p> <p>Oleg Drokin (oleg.drokin@intel.com) merged in patch <a href="https://review.whamcloud.com/27544/" class="external-link" target="_blank" rel="nofollow noopener">https://review.whamcloud.com/27544/</a><br/> Subject: <a href="https://jira.whamcloud.com/browse/LU-9618" title="Connect readahead to prep_partial_page to improve small (&lt; 1 page) write performance" class="issue-link" data-issue-key="LU-9618"><del>LU-9618</del></a> clio: Use readahead for partial page write<br/> Project: fs/lustre-release<br/> Branch: master<br/> Current Patch Set: <br/> Commit: b7d38ece00135e74a627a9702d6dadff10c3c89d</p> <p>Patrick</p> <p>As you will have seen this has landed to master. I saw your most recent musings in gerrit and wondered if you are ok with this or would prefer this reverted while you think some more?</p> <p>Peter</p> <p>Ah, no, it's fine. It's just a little complicated - <a href="https://jira.whamcloud.com/browse/LU-8964" title="use parallel I/O to improve performance on machines with slow single thread performance" class="issue-link" data-issue-key="LU-8964"><del>LU-8964</del></a> removes the readahead code in Lustre, which this patch uses. Once <a href="https://jira.whamcloud.com/browse/LU-8964" title="use parallel I/O to improve performance on machines with slow single thread performance" class="issue-link" data-issue-key="LU-8964"><del>LU-8964</del></a> lands, either as part of that landing or later, the code from this patch will have to be modified to use the kernel readahead infrastructure so we can keep this feature. My musing was about how best to do that. (But we can't make that change to this patch until <a href="https://jira.whamcloud.com/browse/LU-8964" title="use parallel I/O to improve performance on machines with slow single thread performance" class="issue-link" data-issue-key="LU-8964"><del>LU-8964</del></a> lands.)</p> <p>Does that make sense? It's probably on Dmitry to fix it as part of landing <a href="https://jira.whamcloud.com/browse/LU-8964" title="use parallel I/O to improve performance on machines with slow single thread performance" class="issue-link" data-issue-key="LU-8964"><del>LU-8964</del></a> now, but I can help too as desired.</p> <p>Short version:<br/> This patch introduces an additional user of code <a href="https://jira.whamcloud.com/browse/LU-8964" title="use parallel I/O to improve performance on machines with slow single thread performance" class="issue-link" data-issue-key="LU-8964"><del>LU-8964</del></a> will remove. So the code this patch changed will have to be modified when <a href="https://jira.whamcloud.com/browse/LU-8964" title="use parallel I/O to improve performance on machines with slow single thread performance" class="issue-link" data-issue-key="LU-8964"><del>LU-8964</del></a> lands. My musing was on how to do that.</p> <p>ok sounds good. Your review of the <a href="https://jira.whamcloud.com/browse/LU-8964" title="use parallel I/O to improve performance on machines with slow single thread performance" class="issue-link" data-issue-key="LU-8964"><del>LU-8964</del></a> patch would be much appreciated.</p> <p>Minh Diep (minh.diep@intel.com) uploaded a new patch: <a href="https://review.whamcloud.com/31012" class="external-link" target="_blank" rel="nofollow noopener">https://review.whamcloud.com/31012</a><br/> Subject: <a href="https://jira.whamcloud.com/browse/LU-9618" title="Connect readahead to prep_partial_page to improve small (&lt; 1 page) write performance" class="issue-link" data-issue-key="LU-9618"><del>LU-9618</del></a> clio: Use readahead for partial page write<br/> Project: fs/lustre-release<br/> Branch: b2_10<br/> Current Patch Set: 1<br/> Commit: bf77ac10092f22bfbae5873cc3c87425ccb8444c</p> Related LU-11140 LU-8964 LU-12718 Development Rank 1|hzzej3: Rank (Obsolete) 9223372036854775807 Severity