RE: [AMBER] mmpbsa-job-killed

From: Rilei Yu <>
Date: Sat, 10 Apr 2010 07:57:49 +0800 (CST)

Dear Mr. Ross,

Very thanks for your detailed reply! Now, i have more space for the MMPBSA, and it works well. Before I just allocated 2G for the MMMPBSA. But when I changed it to 3 G. Now, everything is ok!

Very thanks again!

Rilei Yu

--- 10年4月9日,周五, Ross Walker <> 写道:

发件人: Ross Walker <>
主题: RE: [AMBER] mmpbsa-job-killed
收件人: "'AMBER Mailing List'" <>
日期: 2010年4月9日,周五,下午1:30

Hi Rilei,

> when I tried to calculate the binding energy
> using the cluster, the job is killed. And when I checked the job file,
> I found the errors as follows:
> =>> PBS: job killed: per node vmem 2254180kb exceeded limit 2048000kb
> I performed the gb and PB on 100 frames. Do you have any idea?

It looks like your job was killed because the node ran out of virtual memory. The limit appears to be 2GB and your job needed more. This is strange though since a MMPBSA calculation shouldn't need that much, unless you are running nmode as part of it on a large system. How far did the calculation get before it died? Did it die on frame 1 or some point after processing a series of frames? The second case would point to some kind of memory leak which is entirely possible but might be tricky to track down. The first case would likely imply either an incorrect setting somewhere, maybe a crazy fine grid or similar setting / typo in a setting. Alternatively you may be trying to run this on a VERY large system.

It would help to have more details about your simulation in order to debug this further.

All the best

|\oss Walker

| Assistant Research Professor |
| San Diego Supercomputer Center |
| Tel: +1 858 822 0854 | EMail:- |
| | |

Note: Electronic Mail is not secure, has no guarantee of delivery, may not be read every day, and should not be used for urgent or sensitive issues. 

AMBER mailing list

AMBER mailing list
Received on Fri Apr 09 2010 - 17:00:04 PDT
Custom Search