Re: [Gluster-devel] Performance experiments with io-stats translator
Nice work! What is the network interconnect bandwidth? How much of the network bandwidth is in use while the test is being run? Wondering if there is saturation in the network layer. -Vijay On Tue, Jun 6, 2017 at 7:35 AM, Krutika Dhananjaywrote: > Hi, > > As part of identifying performance bottlenecks within gluster stack for VM > image store use-case, I loaded io-stats at multiple points on the client > and brick stack and ran randrd test using fio from within the hosted vms in > parallel. > > Before I get to the results, a little bit about the configuration ... > > 3 node cluster; 1x3 plain replicate volume with group virt settings, > direct-io. > 3 FUSE clients, one per node in the cluster (which implies reads are > served from the replica that is local to the client). > > io-stats was loaded at the following places: > On the client stack: Above client-io-threads and above protocol/client-0 > (the first child of AFR). > On the brick stack: Below protocol/server, above and below io-threads and > just above storage/posix. > > Based on a 60-second run of randrd test and subsequent analysis of the > stats dumped by the individual io-stats instances, the following is what I > found: > > *Translator Position* *Avg Latency of READ fop as > seen by this translator* > > 1. parent of client-io-threads1666us > > ∆ (1,2) = 50us > > 2. parent of protocol/client-01616us > > ∆ (2,3) = 1453us > > - end of client stack - > - beginning of brick stack --- > > 3. child of protocol/server 163us > > ∆ (3,4) = 7us > > 4. parent of io-threads156us > > ∆ (4,5) = 20us > > 5. child-of-io-threads 136us > > ∆ (5,6) = 11us > > 6. parent of storage/posix 125us > ... > end of brick stack > > So it seems like the biggest bottleneck here is a combination of the > network + epoll, rpc layer? > I must admit I am no expert with networks, but I'm assuming if the client > is reading from the local brick, then > even latency contribution from the actual network won't be much, in which > case bulk of the latency is coming from epoll, rpc layer, etc at both > client and brick end? Please correct me if I'm wrong. > > I will, of course, do some more runs and confirm if the pattern is > consistent. > > -Krutika > > ___ > Gluster-devel mailing list > Gluster-devel@gluster.org > http://lists.gluster.org/mailman/listinfo/gluster-devel > ___ Gluster-devel mailing list Gluster-devel@gluster.org http://lists.gluster.org/mailman/listinfo/gluster-devel
Re: [Gluster-devel] Performance experiments with io-stats translator
On Tue, Jun 6, 2017 at 5:05 PM, Krutika Dhananjaywrote: > Hi, > > As part of identifying performance bottlenecks within gluster stack for VM > image store use-case, I loaded io-stats at multiple points on the client > and brick stack and ran randrd test using fio from within the hosted vms in > parallel. > > Before I get to the results, a little bit about the configuration ... > > 3 node cluster; 1x3 plain replicate volume with group virt settings, > direct-io. > 3 FUSE clients, one per node in the cluster (which implies reads are > served from the replica that is local to the client). > > io-stats was loaded at the following places: > On the client stack: Above client-io-threads and above protocol/client-0 > (the first child of AFR). > On the brick stack: Below protocol/server, above and below io-threads and > just above storage/posix. > > Based on a 60-second run of randrd test and subsequent analysis of the > stats dumped by the individual io-stats instances, the following is what I > found: > > *Translator Position* *Avg Latency of READ fop as > seen by this translator* > > 1. parent of client-io-threads1666us > > ∆ (1,2) = 50us > > 2. parent of protocol/client-01616us > > ∆ (2,3) = 1453us > > - end of client stack - > - beginning of brick stack --- > > 3. child of protocol/server 163us > > ∆ (3,4) = 7us > > 4. parent of io-threads156us > > ∆ (4,5) = 20us > > 5. child-of-io-threads 136us > > ∆ (5,6) = 11us > > 6. parent of storage/posix 125us > ... > end of brick stack > > So it seems like the biggest bottleneck here is a combination of the > network + epoll, rpc layer? > I must admit I am no expert with networks, but I'm assuming if the client > is reading from the local brick, then > even latency contribution from the actual network won't be much, in which > case bulk of the latency is coming from epoll, rpc layer, etc at both > client and brick end? Please correct me if I'm wrong. > > I will, of course, do some more runs and confirm if the pattern is > consistent. > > -Krutika > > Really interesting numbers! How many concurrent requests are in flight in this test? Could you post the fio job? I'm wondering if/how these latency numbers change if you reduce the number of concurrent requests. -- Manoj ___ Gluster-devel mailing list Gluster-devel@gluster.org http://lists.gluster.org/mailman/listinfo/gluster-devel
[Gluster-devel] Release 3.11.1: Scheduled for 20th of June
Hi, It's time to prepare the 3.11.1 release, which falls on the 20th of each month [4], and hence would be June-20th-2017 this time around. This mail is to call out the following, 1) Are there any pending *blocker* bugs that need to be tracked for 3.11.1? If so mark them against the provided tracker [1] as blockers for the release, or at the very least post them as a response to this mail 2) Pending reviews in the 3.11 dashboard will be part of the release, *iff* they pass regressions and have the review votes, so use the dashboard [2] to check on the status of your patches to 3.11 and get these going 3) Empty release notes are posted here [3], if there are any specific call outs for 3.11 beyond bugs, please update the review, or leave a comment in the review, for us to pick it up Thanks, Shyam/Kaushal [1] Release bug tracker: https://bugzilla.redhat.com/show_bug.cgi?id=glusterfs-3.11.1 [2] 3.11 review dashboard: https://review.gluster.org/#/projects/glusterfs,dashboards/dashboard:3-11-dashboard [3] Release notes WIP: https://review.gluster.org/17480 [4] Release calendar: https://www.gluster.org/community/release-schedule/ ___ Gluster-devel mailing list Gluster-devel@gluster.org http://lists.gluster.org/mailman/listinfo/gluster-devel
[Gluster-devel] Coverity covscan for 2017-06-06-4dba0d5f (master branch)
GlusterFS Coverity covscan results are available from http://download.gluster.org/pub/gluster/glusterfs/static-analysis/master/glusterfs-coverity/2017-06-06-4dba0d5f ___ Gluster-devel mailing list Gluster-devel@gluster.org http://lists.gluster.org/mailman/listinfo/gluster-devel
[Gluster-devel] Performance experiments with io-stats translator
Hi, As part of identifying performance bottlenecks within gluster stack for VM image store use-case, I loaded io-stats at multiple points on the client and brick stack and ran randrd test using fio from within the hosted vms in parallel. Before I get to the results, a little bit about the configuration ... 3 node cluster; 1x3 plain replicate volume with group virt settings, direct-io. 3 FUSE clients, one per node in the cluster (which implies reads are served from the replica that is local to the client). io-stats was loaded at the following places: On the client stack: Above client-io-threads and above protocol/client-0 (the first child of AFR). On the brick stack: Below protocol/server, above and below io-threads and just above storage/posix. Based on a 60-second run of randrd test and subsequent analysis of the stats dumped by the individual io-stats instances, the following is what I found: *Translator Position* *Avg Latency of READ fop as seen by this translator* 1. parent of client-io-threads1666us ∆ (1,2) = 50us 2. parent of protocol/client-01616us ∆ (2,3) = 1453us - end of client stack - - beginning of brick stack --- 3. child of protocol/server 163us ∆ (3,4) = 7us 4. parent of io-threads156us ∆ (4,5) = 20us 5. child-of-io-threads 136us ∆ (5,6) = 11us 6. parent of storage/posix 125us ... end of brick stack So it seems like the biggest bottleneck here is a combination of the network + epoll, rpc layer? I must admit I am no expert with networks, but I'm assuming if the client is reading from the local brick, then even latency contribution from the actual network won't be much, in which case bulk of the latency is coming from epoll, rpc layer, etc at both client and brick end? Please correct me if I'm wrong. I will, of course, do some more runs and confirm if the pattern is consistent. -Krutika ___ Gluster-devel mailing list Gluster-devel@gluster.org http://lists.gluster.org/mailman/listinfo/gluster-devel