On Wednesday 10 August 2011 12:11 AM, Jesse Stroik wrote:
Pavan,
Thank you for your help. We wanted to get back to you with our results
and observations. I'm cc'ing gluster-users for posterity.
We did experiment with enable-trickling-writes. That was one of the
translator tunables we wanted to know the precise syntax for so that we
could be certain we were disabling it. As hoped, disabling trickling
writes improved performance somewhat.
We are definitely interested in any other undocumented write-buffer
related tunables. We've tested the documented tuning parameters.
Performance improved significantly when we switched clients to mainline
kernel (2.5.35-13). We also updated to OFED 1.5.3 but it wasn't
responsible for the performance improvement.
Our findings with 32KB block size (cp) write performance:
250-300MB/sec single stream performance
400MB/sec multiple-stream per client performance
Ok. Lets see if we can improve this further. Please use the following
tunables as suggested below:
For write-behind -
option cache-size 16MB
For read-ahead -
option page-count 16
For io-cache -
option cache-size 64MB
You will need to place these lines in the client volume file, restart
the server and remount the volume on the clients.
Your client (fuse) volume file sections will look like below (of course,
with change in the volume name) -
volume testvol-write-behind
type performance/write-behind
option cache-size 16MB
subvolumes testvol-client-0
end-volume
volume testvol-read-ahead
type performance/read-ahead
option page-count 16
subvolumes testvol-write-behind
end-volume
volume testvol-io-cache
type performance/io-cache
option cache-size 64MB
subvolumes testvol-read-ahead
end-volume
Run your copy command with these tunables. For now, lets have the
default setting for trickling writes which is 'ENABLED'. You can simply
remove this tunable from the volume file to get the default behaviour.
Pavan
This is much higher than we observed with kernel 2.6.18 series. Using
the 2.6.18 line, we also observed virtually no difference between
running single stream tests and multi stream tests suggesting a
bottleneck with the fabric.
Both 2.6.18 and 2.6.35-13 performed very well (about 600MB/sec) when
writing 128KB blocks.
When I disabled write-behind on the 2.6.18 series of kernels as a test,
performance plummeted to a few MB/sec when writing blocks sizes smaller
than 128KB. We did not test this extensively.
Disabling enable-trickling-writes gave us approximately a 20% boost,
reflected in the numbers above, for single-stream writes. We observed no
significant difference with several streams per client due to disabling
that tunable.
For reference, we are running another cluster file system on the same
underlying hardware/software. With both the old kernel (2.6.18.x) and
the new kernel (2.6.35-13) we get approximately:
450-550MB/sec single stream performance
1200MB+/sec multiple stream per client performance
We set the test directory to write entire files to a single LUN which is
how we configured gluster in an effort to mitigate differences.
It is treacherous to speculate why we might be more limited with gluster
over RDMA than the other cluster file system without spending a
significant amount of analysis. That said, I wonder if there may be an
issue with the way in which fuse handles write buffers causing a
bottleneck for RMDA.
The bottom line is that our observed performance was poor using the
2.6.18 RHEL 5 kernel line relative to the mainline (2.6.35) kernels.
Updating to the newer kernels was well worth the testing and downtime.
Hopefully this information can help others.
Best,
Jesse Stroik
_______________________________________________
Gluster-users mailing list
Gluster-users@gluster.org
http://gluster.org/cgi-bin/mailman/listinfo/gluster-users
