Re: size/filesize - or how to initiate a looong run

On Thu, May 24 2007, peter wrote:
> Hello,
>
> When using the following job:
>
> [global]
> filesize=128m
> size=128g
>
> [job1]
> rw=rw
> verify=md5
> randrepeat=1
> fadvise_hint=0
> directory=/var/tmp
> nrfiles=10
> file_service_type=random
>
> I thought that fio would do a total of 128GB of I/O scattered over 10
> files of 128MB each. However, it seems to only do 128*10 = 1280MB of
> I/O. The documentation is a little vague in this area. It often uses the
> word "may", which makes the real intention of size a little vague as well.

I agree, and if you find such weak spots in the documentation, please do
point them out to me so I can fix them!

The above doesn't have any real meaning right now, you would need to
specify a number of loops to reach the 128g. But I think it should work
as you expected it to, it seems the most logical way to handle it. I'll
take a look at fixing that up.

> In order to shed some light on what I want to do: I want to test the iet
> iscsi target with lots of I/O (and I mean lots: It should be a test
> running for many days to see if the implementation is stable enough for
> real use). I plan to either use the rw or randrw I/O patterns with many
> files, a large overall I/O and many syncs. Ultimately, I want to test
> this under XEN with memory pressure while doing a live migration from
> one XEN host to another.

You could just use the 'time_based' option along with a runtime value to
force it to run the workload for as long as you want. Be sure to use the
very latest fio snapshot for that though, a bug in that option was just
reported and fixed yesterday.

> If there is any advise on a useful test setup for such a scenario, I
> would love to hear about it.

I think your job file is a good start, though if you want some syncs,
you need to use the fsync* options.

-- 
Jens Axboe