[Halld-offline] Fwd: [Jlab-scicomp-briefs] JLab data analysis nodes idle; more CentOS 6.2 updates

Mark M. Ito marki at jlab.org
Tue Jul 9 08:36:08 EDT 2013

fyi, from Sandy Philpott

-------- Original Message --------
Subject: 	[Jlab-scicomp-briefs] JLab data analysis nodes idle; more 
CentOS 6.2 updates
Date: 	Mon, 8 Jul 2013 16:05:49 -0400 (EDT)
From: 	Sandy Philpott <philpott at jlab.org>
To: 	jlab-scicomp-briefs at jlab.org

Scientific Computing Users,

With the data analysis cluster currently idle, the 2010 farm nodes will be updated to CentOS 6.2 at this time.  These 192 cores will be added to the 672 cores in the 2011-2012 nodes already running CentOS 6.2.
The 2008-2009 nodes of 280 cores remain at CentOS 5.3.

In addition, the FY13 procurement for compute nodes is in progress, and will add another 448 cores to the cluster. These nodes are expected to arrive this fall.

   Sandy Philpott

----- Original Message -----
From: "Sandy Philpott" <philpott at jlab.org>
To: jlab-scicomp-briefs at jlab.org
Sent: Tuesday, May 14, 2013 3:01:50 PM
Subject: JLab data analysis nodes available


Following the recent 10-day lull in batch farm computing for data analysis, the queues are now returning to full. It's a good time to remind users that the farm is migrating from CentOS 5.3 to 6.2 this summer, so it's time to make sure your jobs will run on the 6.2 systems.  Currently all the 2011 and 2012 systems are running CentOS 6.2; the older systems from 2008-2010 are still at 5.3 for now.  The https://scicomp.jlab.org web page has been updated to show the number of nodes and percent usage of each type.

The 6.2 interactive node is "ifarm.jlab.org", currently host "ifarm1102". Use the tag "OS: centos62" to submit jobs to the CentOS 6.2 farm systems.

It's also a good idea to revisit the resource requirements you include in your job submission, to make sure you request only the extra memory, time, or disk space that's needed. You'll find the actual usage numbers in your ~/.farm_out/*out files after your jobs complete.

Some groups are currently using less than their fairshare, allowing other groups a chance for more processing. Current allocations are set at
   Hall A:      10%
   Hall B/CLAS: 50%
   Hall C:      10%
   Hall D:      30%

Usage information is available on the web at https://scicomp.jlab.org/scicomp/#/auger/usage.

   Sandy Philpott

Jlab-scicomp-briefs mailing list
Jlab-scicomp-briefs at jlab.org

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://mailman.jlab.org/pipermail/halld-offline/attachments/20130709/54feb548/attachment-0002.html>

More information about the Halld-offline mailing list