[Pw_forum] davcio error #20 while running epw.x in parallel mode

[Elias Albert]

Dear all,
I am using the EPW package in conjunction with QE to calculate electron phonon coefficients for systems. The "scf", "ph" and "nscf" run quite smoothly; however when I execute the last step which is using the "epw.x" command I get the error:
"Q(r) pseudized with 0 coefficients
     EPW          :     6.72s CPU time,   17.23s wall time
     EPW          :    14.08s CPU time,   28.02s wall time
     EPW          :    14.08s CPU time,   28.02s wall time

 %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%     from setphases_wrap : error #         1     only one proc per pool %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%"
"The parallel execution is such that:
Program EPW       v.2.3.5  starts ...     Today is 12Nov2012 at  2:27:22
     Parallel version (MPI)
     Number of processors in use:      16     R & G space division:  proc/pool =   16"
However when I reduce the number to 8 processors (that is 1 node on the cluster I am working on), I get another error from davcio:
"%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%     from davcio : error #        20     error while reading from file %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%"
Mind you that the previous scf , nscf and ph steps were done with 16 processors. Could that be the error? Would repeating everything with 8 processors work?
Thanks in advance
Elias ALbertUniiversity of DerbyDE12 4ERUK





 		 	   		  
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.quantum-espresso.org/pipermail/users/attachments/20121112/bbf582c3/attachment.html>