[slurm-users] Stagein/Stageout

[John Hearns]

Andrew, I looked at this about a year ago.  You might find the thread in
the archives of this list.
At the time, the Cray plugin for the burst buffer was supported. However
staging in/out for other devices was not being developed.

You can achieve the same staging behaviour by using job dependencies.
Submit a stage-in job. then submit a second job which depends on this job
to do the computation.

On 6 January 2018 at 11:47, Marcin Stolarek <stolarek.marcin at gmail.com>
wrote:

> If nothing changed recently the shared filesystem like nfs/gpfs/lustre is
> a requirement for normal cluster configuration. You can workaround it with
> prologue/spank plugins but  honestly I haven't seen real hpc cluster
> without shared filesystem.
>
> cheers,
> Marcin
>
> 2018-01-05 23:25 GMT+01:00 Andrew Melo <andrew.melo at gmail.com>:
>
>> Hi all,
>>
>> Does SLURM support similar functionality to the PBS options
>>
>> #PBS -w stagein
>> #PBS -w stageout
>>
>> Looking through the docs and even the qsub wrapper commands, I don't
>> see an analogous way to implement the same with SLURM via sbatch. I
>> see the documentation about burst buffers, but that doesn't seem to be
>> the same.
>>
>> Any suggestions?
>>
>> Andrew
>>
>> --
>> --
>> Andrew Melo
>>
>>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.schedmd.com/pipermail/slurm-users/attachments/20180106/e52b1540/attachment.html>