forum@abinit.org
Subject: The ABINIT Users Mailing List ( CLOSED )
List archive
- From: Adam Sorini <asorini@gmail.com>
- To: forum@abinit.org
- Subject: Re: [abinit-forum] parallelization
- Date: Wed, 18 Nov 2009 09:19:47 -0800
- Domainkey-signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=mime-version:in-reply-to:references:date:message-id:subject:from:to :content-type; b=K53tou44uJ6jX/C/FCKajZ4njq5AL8UEaslxMsWNWb3qVK1FGPtD3ObpNM5hAMn++L aB950sraPuFWLIBdFq366Jl1j/jHkI/mhtgeXoXR6IPMOUtjKKY8wRD8OGIb/sIIBkms SV1Ih4DBYW/BZ6BLjmMSQHK2BY7zxpvz20VhQ=
Hi,
Regarding your first question, the tutorials are probably old and so they don't utilize the more new-fangled "kgb" parallelization. Regarding your second question... you need the ONLY k-point parallelization because after your paral_kgb=1 calculation crashes you can switch to parallelization over ONLY k-points. Cheers,
Adam
On Wed, Nov 18, 2009 at 7:29 AM, <mohua12@uga.edu> wrote:
Hi all,
I am absolutely new to abinit and my apologies if my questions are trivial.
I was looking at the parallelization examples tparal_1.in tparal_2.in and
tparal_3.in and was expecting to see parallelization variables like paral_kgb,
npfft, npband, npkpt, fft_opt_lob and others. I did not see any and still
parallelization is activated.
Could anyone please tell me what am I missing here?
The other question is since paral_kgb = 1 activates parallelization over bands,
FFTs, and k-point/spin what do I need to do for ONLY k-point parallelization
for non-spin calculation.
Thanks
Mohua Bhattacharya
--
Adam Sorini
Post Doctoral Scholar
Stanford Institute for Materials and Energy Science (SIMES)
SLAC National Accelerator Center
2575 Sand Hill Rd., MS 69
Menlo Park, CA 94025
asorini@stanford.edu
- [abinit-forum] parallelization, mohua12, 11/18/2009
- Re: [abinit-forum] parallelization, Adam Sorini, 11/18/2009
Archive powered by MHonArc 2.6.16.