Description of problem: We saw improvement in IOPs as part of performance improvement effort for gluster with vm store use case. As per volume profile data collected, qemu sends almost as many stats as the number of reads in pure random read workload. With stat-prefetch on, the number of stats hitting the disk got reduced to < 100. Version-Release number of selected component (if applicable): How reproducible: Steps to Reproduce: 1. 2. 3. Actual results: Expected results: Additional info:
REVIEW: https://review.gluster.org/17713 (extras: Enable stat-prefetch in virt profile) posted (#1) for review on master by Krutika Dhananjay (kdhananj)
COMMIT: https://review.gluster.org/17713 committed in master by Pranith Kumar Karampuri (pkarampu) ------ commit f28b5d1007644749761fd374d951b72ace2e203e Author: Krutika Dhananjay <kdhananj> Date: Thu Jul 6 14:25:09 2017 +0530 extras: Enable stat-prefetch in virt profile In the internal testing that was done, stat-prefetch did help reduce the number of stats coming from qemu hitting the disk, and thereby improved performance. Change-Id: Icf1ce62ecf4e96b97e1946a77b30434157a7786a BUG: 1468191 Signed-off-by: Krutika Dhananjay <kdhananj> Reviewed-on: https://review.gluster.org/17713 CentOS-regression: Gluster Build System <jenkins.org> Smoke: Gluster Build System <jenkins.org> Reviewed-by: Amar Tumballi <amarts> Reviewed-by: Pranith Kumar Karampuri <pkarampu>
This bug is getting closed because a release has been made available that should address the reported issue. In case the problem is still not fixed with glusterfs-3.12.0, please open a new bug report. glusterfs-3.12.0 has been announced on the Gluster mailinglists [1], packages for several distributions should become available in the near future. Keep an eye on the Gluster Users mailinglist [2] and the update infrastructure for your distribution. [1] http://lists.gluster.org/pipermail/announce/2017-September/000082.html [2] https://www.gluster.org/pipermail/gluster-users/