Hi Marc,
We actually
kept the benchmark running through an upgrade from mimic to
octopus. Was quite interesting to see how certain performance properties
change with that.
So you have stats that show the current performance of a host having mimic and from
another host that has octopus?
I don't know what you mean by that. We have a kclient ceph fs mount to the cluster and
we upgrade the ceph cluster while the benchmark is running. The kclient is on a kernel
version, not a ceph version and we upgrade the ceph cluster, not the client OS.
This benchmark
makes it possible to compare versions
with live timings coming in.
Do you do something with this time output like store it seperately in prometheus/influx?
Or all your statistics coming from what is being reported by ceph itself?
The output is tailored for grep. The script bench-grep has a list of useful expressions to
search for. If you use this suite, you should aim for each benchmark taking about 30
minutes. Its not a lot of numbers per day and I just read through them. Even if you run
this for a few weeks, the grep output is quite readable.
If you want to do something fancy, knock yourself out :)
Best regards,
=================
Frank Schilder
AIT Risø Campus
Bygning 109, rum S14
________________________________________
From: Marc <Marc(a)f1-outsourcing.eu>
Sent: Wednesday, May 31, 2023 12:37 PM
To: Frank Schilder
Cc: ceph-users(a)ceph.io
Subject: RE: [ceph-users] Re: CEPH Version choice
Hi Frank,
Thanks! I have added this to my test environment todo
I uploaded all scripts and a rudimentary readme to
https://github.com/frans42/cephfs-bench . I hope it is sufficient to get
started. I'm afraid its very much tailored to our deployment and I can't
make it fully configurable anytime soon. I hope it serves a purpose
though - at least I discovered a few bugs with it.
I think I know where I should add something for creating and comparing hashes. For me this
integrity of el9 + Nautilus is most important.
We actually kept the benchmark running through an
upgrade from mimic to
octopus. Was quite interesting to see how certain performance properties
change with that.
So you have stats that show the current performance of a host having mimic and from
another host that has octopus?
This benchmark makes it possible to compare versions
with live timings coming in.
Do you do something with this time output like store it seperately in prometheus/influx?
Or all your statistics coming from what is being reported by ceph itself?
ps. my el7/el9 time does not have an option -f