REVIEW: http://review.gluster.org/5897 (gfapi: use native STACK_WIND for read _async() calls) posted (#4) for review on master by Anand Avati (avati)
COMMIT: http://review.gluster.org/5897 committed in master by Anand Avati (avati) ------ commit 8eb3898578a4fe359934da57c0e51cfaa21685d4 Author: Anand Avati <avati> Date: Wed Sep 11 00:49:57 2013 -0700 gfapi: use native STACK_WIND for read _async() calls There is little value in using synctask wrappers for async IO requests, as STACK_WIND is asynchronous by nature already. Skip going through synctask for read/write async calls. Change-Id: Ifde331d7c97e0f33426da6ef4377c5ba70dddc06 BUG: 1009134 Signed-off-by: Anand Avati <avati> Reviewed-on: http://review.gluster.org/5897
Did this change get committed to nightly builds for RHS 2.1U2? When? This result looks like single-thread read performance before the patch (and after stat-prefetch on tuning applied) [root@gprfc093 ~]# rpm -q glusterfs glusterfs-3.4.0.35.1u2rhs-1.el6rhs.x86_64 [root@gprfc093 ~]# ssh gprfs047 rpm -q glusterfs glusterfs-3.4.0.35.1u2rhs-1.el6rhs.x86_64 [root@gprfc093 ~]# ssh gprfs048 rpm -q glusterfs [root@gprfc093 ~]# virsh create /home/kvm_images/gfapi.xml Domain gfapi created from /home/kvm_images/gfapi.xml [root@gprfs047 ~]# gluster v i Volume Name: benvol Type: Replicate Volume ID: acbd0a41-35ee-466a-99bb-07a0cfe7bc81 Status: Started Number of Bricks: 1 x 2 = 2 Transport-type: tcp Bricks: Brick1: gprfs047-10ge:/mnt/b1/brick Brick2: gprfs048-10ge:/mnt/brick0/brick Options Reconfigured: cluster.self-heal-daemon: off network.remote-dio: enable cluster.eager-lock: enable performance.io-cache: off performance.read-ahead: off performance.quick-read: off server.allow-insecure: on diagnostics.count-fop-hits: on diagnostics.latency-measurement: on performance.stat-prefetch: on [root@gprfc093 ~]# for n in 7 8 ; do ssh gprfs04$n 'echo 1 > /proc/sys/vm/drop_caches' ; done [root@gprfc093 ~]# ssh gfapi 'echo 1 > /proc/sys/vm/drop_caches ; iozone -w -c -e -i 1 -+n -r 64k -s 16g -t 1 -F /mnt/vbd-b/f.tmp /mnt/vbd-c/f.tmp /mnt/vbd-d/f.tmp /mnt/vbd-e/f.tmp ' Children see throughput for 1 readers = 581969.38 KB/sec Parent sees throughput for 1 readers = 581762.98 KB/sec
COMMIT: http://review.gluster.org/6325 committed in release-3.4 by Anand Avati (avati) ------ commit 24e4b5d12be5d92a4e5c3167372f88cd3dfa720a Author: Anand Avati <avati> Date: Wed Sep 11 00:49:57 2013 -0700 gfapi: use native STACK_WIND for read _async() calls There is little value in using synctask wrappers for async IO requests, as STACK_WIND is asynchronous by nature already. Skip going through synctask for read/write async calls. Change-Id: Ifde331d7c97e0f33426da6ef4377c5ba70dddc06 BUG: 1009134 Signed-off-by: Anand Avati <avati> Reviewed-on: http://review.gluster.org/6325 Tested-by: Gluster Build System <jenkins.com>
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-3.4.3, please reopen this bug report. glusterfs-3.4.3 has been announced on the Gluster Developers mailinglist [1], packages for several distributions should already be or become available in the near future. Keep an eye on the Gluster Users mailinglist [2] and the update infrastructure for your distribution. The fix for this bug likely to be included in all future GlusterFS releases i.e. release > 3.4.3. In the same line the recent release i.e. glusterfs-3.5.0 [3] likely to have the fix. You can verify this by reading the comments in this bug report and checking for comments mentioning "committed in release-3.5". [1] http://thread.gmane.org/gmane.comp.file-systems.gluster.devel/5978 [2] http://news.gmane.org/gmane.comp.file-systems.gluster.user [3] http://thread.gmane.org/gmane.comp.file-systems.gluster.devel/6137
This bug has been CLOSED, and there has not been a response to the requested NEEDINFO in more than 4 weeks. The NEEDINFO flag is now getting cleared so that our Bugzilla household is getting more in order.