<div dir="ltr"><div>Andrew, I looked at this about a year ago. You might find the thread in the archives of this list.</div><div>At the time, the Cray plugin for the burst buffer was supported. However staging in/out for other devices was not being developed.</div><div><br></div><div>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.</div></div><div class="gmail_extra"><br><div class="gmail_quote">On 6 January 2018 at 11:47, Marcin Stolarek <span dir="ltr"><<a href="mailto:stolarek.marcin@gmail.com" target="_blank">stolarek.marcin@gmail.com</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><div dir="ltr"><div><div>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.<br><br></div>cheers,<br></div>Marcin<br></div><div class="HOEnZb"><div class="h5"><div class="gmail_extra"><br><div class="gmail_quote">2018-01-05 23:25 GMT+01:00 Andrew Melo <span dir="ltr"><<a href="mailto:andrew.melo@gmail.com" target="_blank">andrew.melo@gmail.com</a>></span>:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">Hi all,<br>
<br>
Does SLURM support similar functionality to the PBS options<br>
<br>
#PBS -w stagein<br>
#PBS -w stageout<br>
<br>
Looking through the docs and even the qsub wrapper commands, I don't<br>
see an analogous way to implement the same with SLURM via sbatch. I<br>
see the documentation about burst buffers, but that doesn't seem to be<br>
the same.<br>
<br>
Any suggestions?<br>
<span class="m_-7007667239868684563HOEnZb"><font color="#888888"><br>
Andrew<br>
<br>
--<br>
--<br>
Andrew Melo<br>
<br>
</font></span></blockquote></div><br></div>
</div></div></blockquote></div><br></div>