[Pw_forum] ph.x high I/O frequency

[Jiseok Kim]

Dear, QE-users

Recently, I have received complaints from server administrators as the ph.x
causing excessively high I/Os.
As far as I understand, the ph.x generates *.bar.XXX, *.dwfXXX, *.midxXXX,
*.recoverXXX and *.wfcXXX under _ph0 directory where the 'XXX' is the number
of cores.
The server admin complains that the default setting for ph.x
(reduce_io=.false.) is extremely conservative and saves too often affecting
other users.
So I tried to reduce the I/O by setting the option 'reduce_io=.true.'.
Then I could not restart the job from the interruption using
'recover=.true.'.
(Of course I could restart from the 'reduce_io=.false.'.)

Could anyone share your experience how to reduce the I/O ?
I'm trying to run 4 group of q-points using 240 cores for each.
So I'm trying to run using total 960 cores which generates and writes total
5*960=4800 files under _ph0.

Thank you.

JISEOK KIM, Ph.D.
Postdoctoral Research Associate
Department of Materials Science and Engineering
University of Texas at Dallas
800 W. Campbell Rd. RL10
Richardson, TX 75080
(413)386-6285


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.quantum-espresso.org/pipermail/users/attachments/20121016/b7b42ba1/attachment.html>