<html>
<head>
<style><!--
.hmmessage P
{
margin:0px;
padding:0px
}
body.hmmessage
{
font-size: 10pt;
font-family:Tahoma
}
--></style></head>
<body class='hmmessage'><div dir='ltr'>
Hello,<div><br></div><div>It is not a computer. It is a cluster and I am doing parallel computing on 3 nodes each with 8 processors which is a total of 24 processors; this is why I am surprised that the calculation is slow . I tried performing the calculations on different nodes but in vain. </div><div><br></div><div>N.B: I did calculations on one server with only 8 processors for the Gamma point and the calculations finished in three weeks (though with some negative frequencies, this is why I decided to minimize the tr2_ph) But now that server is down. I thought on more nodes things would be faster; apparently not</div><div><br></div><div><br></div><div>Regards</div><div><br></div><div>Elie <br><br><div><div id="SkyDrivePlaceholder"></div>> Date: Mon, 19 Mar 2012 10:00:37 -0400<br>> Subject: Re: [Pw_forum] parallel computation too slow<br>> From: akohlmey@gmail.com<br>> To: elie.moujaes@hotmail.co.uk<br>> CC: pw_forum@pwscf.org<br>> <br>> On Mon, Mar 19, 2012 at 9:45 AM, Elie M <elie.moujaes@hotmail.co.uk> wrote:<br>> > Professor Giannozzi,<br>> ><br>> > Thanks for the reply. The CPU time is 1 hr 26 min (2hrs 2 min normal time)<br>> > for the scf calculation on the nodes (3 of them). If I have 42 atoms then<br>> > this is equivalent to almost 8 days CPU time or 10 days normal time. Four<br>> > days have passed and I am still in representation 6 out of 126 which i guess<br>> > is not the normal situation...Can anything be done (or checked) in this<br>> > case?<br>> <br>> get a real computer! if you already lose a quarter of the time due to<br>> swapping for the scf calculation, you'll be seriously out of memory<br>> for anything more advanced. your laptop is obviously not suitable<br>> to carry out this kind of calculation. sorry.<br>> <br>> axel.<br>> <br>> ><br>> > Regards<br>> ><br>> ><br>> > Elie<br>> ><br>> >> CC: pw_forum@pwscf.org<br>> >> From: giannozz@democritos.it<br>> >> Subject: Re: [Pw_forum] parallel computation too slow<br>> >> Date: Fri, 16 Mar 2012 21:48:15 +0100<br>> >> To: elie.moujaes@hotmail.co.uk<br>> ><br>> >><br>> >><br>> >> On Mar 16, 2012, at 19:14 , Elie M wrote:<br>> >><br>> >> > I am running phonon calculations for a supercell of 60 atoms at the<br>> >> > gamma point [...]<br>> >> > What could the possible causes of this slowness be? Is it something<br>> >> > to do with the<br>> >> > nodes or with phonon calculations in general? Is there a way to check?<br>> >><br>> >> ><br>> >><br>> >> a phonon calculation for N atoms requires a CPU time of the order of<br>> >> 3N the time of<br>> >> the correspoding scf calculation. Does yours? if it does, there is<br>> >> nothing that can be<br>> >> done.<br>> >><br>> >> P.<br>> >> ---<br>> >> Paolo Giannozzi, Dept of Chemistry&Physics&Environment,<br>> >> Univ. Udine, via delle Scienze 208, 33100 Udine, Italy<br>> >> Phone +39-0432-558216, fax +39-0432-558222<br>> >><br>> >><br>> >><br>> >><br>> ><br>> > _______________________________________________<br>> > Pw_forum mailing list<br>> > Pw_forum@pwscf.org<br>> > http://www.democritos.it/mailman/listinfo/pw_forum<br>> ><br>> <br>> <br>> <br>> -- <br>> Dr. Axel Kohlmeyer<br>> akohlmey@gmail.com http://goo.gl/1wk0<br>> <br>> College of Science and Technology<br>> Temple University, Philadelphia PA, USA.<br></div></div> </div></body>
</html>