Hi Michael
Pathological is very subjective, but the standard IOPS test with IOR
involves doing 4K writes (or reads) with random but non-overlapping offsets:
srun -N 4 -n 128 ./ior -t 4096 -b 4096 -s 1024 -e -w -z
However IOR is not really designed to do truly terrible I/O patterns such
as those
> > Up to the point when you’ve experienced problems clearing records, had at
> > any point Changelogs been consumed (read/cleared) in parallel?
> Yes indeed!
>
> There was parallel reading and clearing of the change logs. We had both
> robinhood as a daemon reading clearing and attempts to do th
On Jun 1, 2017, at 19:34, Russell Dekema wrote:
>
> Greetings,
>
> Is there a way, either on the Lustre clients or (preferably) OSSes, to
> determine how many I/O operations each Lustre client is performing
> against the filesystem?
>
> I know several ways of finding the number of *bytes* read
Hi Andy,
Yes indeed!
There was parallel reading and clearing of the change logs. We had both
robinhood as a daemon reading clearing and attempts to do the same manually for
the same user at the same time.
We simply didn’t know this would be an issues.
Faye
From: Andy Moe [mailto:m...@cray.co
Hi all,
I am looking for IOR scripts that represent pathological use cases for file
systems. Something like shared file access with a small, unaligned block size
or random I/O to a shared file. Does anyone has some input for me there that
he/she is willing to share?
Regards, Michael
--
Dr.-