[Pw_forum] Calculating multiple el-ph matrices simultaneously

Huiqun Zhou hqzhou at nju.edu.cn
Wed Jun 9 07:17:36 CEST 2010


Carefully read the "Additional information" of INPUT_PH.html, then you 
can do easilly what you want.

Huiqun Zhou
@Earth Sciences, Nanjing University, China
  ----- Original Message ----- 
  From: Hahnbidt Rhee 
  To: pw_forum at pwscf.org 
  Sent: Wednesday, June 09, 2010 8:56 AM
  Subject: [Pw_forum] Calculating multiple el-ph matrices simultaneously


  Hello, all. I am doing electron-phonon calculations, and my question pertains specifically to getting the el-ph matrices for each q-point. I have noticed that, although calculating a matrix of one q-point is independent of that of another q-point, the code calculates one q-point at a time--ie, one after the other--which makes for an inefficient and very time-consuming run. This is based on the elph.in input file (for ph.x), given in Example07 of the examples given by PWSCF, in which there is a line to specify the q-pt grid,


      nq1=4, nq2=4, nq3=4


  but not individual q-pts. So one run takes care of all the q-points. Is there an option (that I don't know of) to create separate jobs for each q-point so that I can calculate el-ph matrices of multiple inequivalent q-pts SIMULTANEOUSLY? I am using the parallel version of the latest code (v4.2), so it seems reasonable for there to be this option. Thank you!

  HB Rhee
  UC Davis



------------------------------------------------------------------------------


  _______________________________________________
  Pw_forum mailing list
  Pw_forum at pwscf.org
  http://www.democritos.it/mailman/listinfo/pw_forum
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://www.democritos.it/pipermail/pw_forum/attachments/20100609/f3f3a047/attachment.htm 


More information about the Pw_forum mailing list