Just a quick follow up - I had to somehow miss the updated thread count
because fio (client side) and brick show appropriate number of threads if I
raise count above 16. Seems like minimal thread count is always spawned at
least on the server which probably confused me. Unfortunately the
connection count to the brick is still the same (2).
However the read perfomance is still the same. strace on fio threads shows
reads (same on the server side), so the workload is somehow distributed
between them, but performance is very different from running multiple jobs
(numjobs > 1). Does anyone know how fio resp. gfapi ioengine uses multiple
threads?
The 200MB/s I get on client vs 1.5GB/s if the same test is run locally on
the server so far seems to be caused by the network latency, but could the
client be advised to open multiple connections (maybe one per thread)?
netstat reports 2 connections per fio process and raising numjobs results
in more connections and maxes out brick utilization.
My goal is to max out IO performance of QEMU/KVM guests. So far only
approximately 200MB/s are possible with for certain block sizes.
-ps
On Mon, Nov 7, 2016 at 4:47 PM, Pavel Szalbot <pavel.szalbot at gmail.com>
wrote:
> Hi everybody,
>
> I am trying to benchmark my cluster with fio's gfapi ioengine and
evaluate
> effect of various volume options on performance. I have so far observed
> following:
>
> 1) *thread* options do not affect performance or thread count - htop
> always show 2 threads on client, there are always 16 glusterfsd threads on
> server
> 2) running the same test locally (on the brick) shows up to 5x better
> throughput than over 10GBe (MTU 9000, iperfed, pinged with DF set, no drops
> on switches or cards, tcpdumped to check network issues)
> 3) performance.cache-size value has no effect on performance (32MB or 1GB)
>
> I would expect raising client threads number leading to more TCP
> connections, higher disk utilization and throughput. If I run multiple fio
> jobs (numjobs=8), I am able to saturate the network link.
>
> Is this normal or I am missing something really badly?
>
> fio config:
>
> [global]
> name=gfapi test
> create_on_open=1
> volume=test3-vol
> brick=gfs-3.san
> ioengine=gfapi
> direct=1
> bs=256k
> rw=read
> iodepth=1
> numjobs=1
> size=8192m
> loops=1
> refill_buffers=1
> [job1]
>
> reconfigured volume options:
> performance.client-io-threads: on
> performance.cache-size: 1GB
> performance.read-ahead: off
> server.outstanding-rpc-limit: 128
> performance.io-thread-count: 16
> server.event-threads: 16
> client.event-threads: 16
> nfs.disable: on
> transport.address-family: inet
> performance.readdir-ahead: on
>
> -ps
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
<http://www.gluster.org/pipermail/gluster-users/attachments/20161108/b2a3009f/attachment.html>