Re: question on the parallel job on NCSA Xeon cluster

From: Gengbin Zheng (gzheng_at_ks.uiuc.edu)
Date: Tue Aug 17 2004 - 23:01:02 CDT

• Next message: Lv Shouqin: "how to get hydrogen-bond energy?"
• Previous message: Hyonseok Hwang: "question on the parallel job on NCSA Xeon cluster"
• In reply to: Hyonseok Hwang: "question on the parallel job on NCSA Xeon cluster"
• Messages sorted by: [ date ] [ thread ] [ subject ] [ author ] [ attachment ]

Hi,

If I remember right, you need to use "-poll" option to cmpirun. It is
explained in the login screen I think.

Gengbin

Hyonseok Hwang wrote:

> Dear everyone,
>
> I'm running parallel NAMD jobs on the Xeon cluster (named
> 'tungsten')at NCSA. When I use a single or two processors, the jobs
> are running OK. But, if I try to use 4 or 8 processors, there is an
> error message saying "segmentation fault" after 4 or 5 steps and jobs
> just quit. In this parallel jobs, I don't use 'charmrun' command,
> instead I use 'cmpirun', which is used in the Xeon cluster.
> I'm wondering if this problem is caused by network configuration or
> something else. If you know anything, let me know, please.
> Thank you very much.
>
> -Hyon

• Next message: Lv Shouqin: "how to get hydrogen-bond energy?"
• Previous message: Hyonseok Hwang: "question on the parallel job on NCSA Xeon cluster"
• In reply to: Hyonseok Hwang: "question on the parallel job on NCSA Xeon cluster"
• Messages sorted by: [ date ] [ thread ] [ subject ] [ author ] [ attachment ]

This archive was generated by hypermail 2.1.6 : Wed Feb 29 2012 - 15:38:50 CST