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-979 Lustre <p>Formatting a lustre filesystem in RHEL6 barriers are enabled by default with our Lustre 2.0 (same for Lustre 2.1). The impact on small IO performance is very important if we don't specify nobarrier as an MDT mount option. A single fileop bench running with '-f 15' will take more than 2m30sec while the one with nobarrier as mount option on MDT will take less than a minute. Compiling a kernel in Lustre is also a good way to see the problem due to jbd2 barriers.</p> <p>This problem had already been addressed in SLES with ldiskfs patch ext4-disable-write-bar-by-default.patch. I adapted a similar patch to RHEL6.1. I'll try to submit it as soon as possible.</p> RHEL6.1 with Lustre2.0/2.1 LU-979

Barriers should be disabled by default on RHEL6

Bug Minor Resolved Duplicate Yang Sheng Diego Moreno ldiskfs performance rhel Tue, 10 Jan 2012 10:03:22 +0000 Fri, 31 May 2013 18:21:33 +0000 Tue, 10 Jan 2012 10:33:52 +0000 Lustre 2.1.0 0 3 <p>I think the latest patchset of <a href="https://jira.whamcloud.com/browse/LU-751" title="Kernel update [RHEL6.2 2.6.32-220.el6]" class="issue-link" data-issue-key="LU-751"><del>LU-751</del></a> (available here <a href="http://review.whamcloud.com/#change,1892" class="external-link" target="_blank" rel="nofollow noopener">http://review.whamcloud.com/#change,1892</a>) already takes care of this.</p> <p>YangSheng</p> <p>Please consider this when working on the RHEL6.2 support</p> <p>Thanks</p> <p>Peter</p> <p>Right Johann,</p> <p>thanks!</p> <p>Ah good - thanks YangSheng\Johann!</p> <p>Hi,</p> <p>Coming back to this issue, I couldn't see in the patch where are you disabling barriers for rhel6.2.</p> <p>Hi,</p> <p>Coming back to this issue, could you please specify where are barriers disabled on lustre 2.1 code? I cannot see anywhere any code disabling barriers by default. This question also stands for lustre 2.2 but in this case the patch is there but it's not applied.</p> <p>Thanks,</p> <p>Hi, Diego, As Andreas point out in gerrit:</p> <div class="preformatted panel" style="border-width: 1px;"><div class="preformattedContent panelContent"> <pre>I think we should leave barriers enabled by default, and if performance is down for specific types of hardware then the administrator can choose to mount with "-o nobarrier" by themselves. I'm not sure this will happen, because the kernel barrier implementation has improved significantly since SLES10 when it was first needed, and it is enabled by default for both ext3 and ext4 in all modern kernels. </pre> </div></div> <p>I think nobarrier option for OSTs really helps to improve direct IO perforamnce. I saw dramatically performance improvements with nobarrier option.<br/> In the direct IO mode, journal transaction starts for each RPC to commit to disk and jbd issues barrier to storage for each journal commit. If there are battery backup raid cache and cache mirroring feature between storage controllers, nobarrier can be enabled and help a lot to reduce the latency for journal commeit. thgouht?</p> Development Rank 1|hzvhjj: Rank (Obsolete) 6491 Severity