From owner-nwchem-users@emsl.pnl.gov Thu Dec 28 02:25:04 2006
Received: from odyssey.emsl.pnl.gov (localhost [127.0.0.1])
	by odyssey.emsl.pnl.gov (8.13.8/8.13.8) with ESMTP id kBSAP3eb010919
	for <nwchem-users-outgoing-0915@odyssey.emsl.pnl.gov>; Thu, 28 Dec 2006 02:25:04 -0800 (PST)
Received: (from majordom@localhost)
	by odyssey.emsl.pnl.gov (8.13.8/8.13.8/Submit) id kBSAP3a9010918
	for nwchem-users-outgoing-0915; Thu, 28 Dec 2006 02:25:03 -0800 (PST)
X-Authentication-Warning: odyssey.emsl.pnl.gov: majordom set sender to owner-nwchem-users@emsl.pnl.gov using -f
X-Ironport-SG: OK_Domains
X-Ironport-SBRS: 2.8
X-IronPort-Anti-Spam-Filtered: true
X-IronPort-Anti-Spam-Result: AgAAAJcpk0VC+VLjgGdsb2JhbACOCQEBCQws
X-IronPort-AV: i="4.12,215,1165219200"; 
   d="scan'208"; a="10787610:sNHT17401286"
DomainKey-Signature: a=rsa-sha1; q=dns; c=nofws;
        s=beta; d=gmail.com;
        h=received:message-id:date:from:sender:to:subject:mime-version:content-type:content-transfer-encoding:content-disposition:x-google-sender-auth;
        b=tY5XSkKk7wsajC7SniIylS7Z9bpwA9CZmrt7mZXVenMOXI1e8wStIJLoIQOCVDcgwUdg28kkBAqaCCHtYcH9XAPL5FwmmghoTsM+HJeZE8vCX054vR1C3Nurny31I2xjwkr0A94d4CLFLEn4ydfBqpo9erczYzkriE64xGdGRwM=
Message-ID: <14da35910612280224t45a43424xccfd4ae5e99a83ef@mail.gmail.com>
Date: Thu, 28 Dec 2006 02:24:58 -0800
From: alex@servermet.com.ru
To: nwchem-users@emsl.pnl.gov
Subject: [NWCHEM] NWCHEN on SMP cluster, Multi-threading question
MIME-Version: 1.0
Content-Type: text/plain; charset=ISO-8859-1; format=flowed
Content-Transfer-Encoding: 7bit
Content-Disposition: inline
X-Google-Sender-Auth: 1f171ad708f7a5cf
Sender: owner-nwchem-users@emsl.pnl.gov
Precedence: bulk

Hi All, I'm running nwchem-5.0 on a cluster where each node has 2
processors (SMP). It appears to me that nwchem automatically invokes
multi-threading and uses both processors if detected.

For example, when I submitted a job, say, to 4 nodes and requested 1
processor per node, I noticed that the job actually ran on both
processors on each node. Is this normal? It would be a problem when
someone else was using the 2nd processor on the same nodes. The job
would still run on two threads and each of thread may consume 50% CPU
usage on 1 processor, or one of the threads would jump on the other
processor and push the CPU usage of other's job down to 50%. Not only
it slowed down the compution by taking too much resources, but also
interfered others' jobs. So, is there a way to control the
multi-threading in nwchem?

Thanks in advance,

-Alex