[Jlab-scicomp-briefs] JLab Data Analysis systems maintenance: Wed Nov 14

Sandy Philpott philpott at jlab.org
Thu Nov 8 15:21:39 EST 2012


Dear Users:

The Scientific Computing Group will perform a major software upgrade on 
Wednesday Nov. 14. During the upgrade, all scicomp services of JasMine 
and Auger will not be available. We are anticipating one day outage of 
scicomp service, but the outage may last into Thursday if we encounter 
some unexpected difficulties.

Additionally, a few operational changes will occur:

* The interactive user node ifarm1102 will be removed from the "ifarml64"
  alias, to be rebuilt at as a CentOS 6.2 "ifarm" node.

* The farm job memory limit will be increased from 4GB to 10GB.

* The ROOT PRO link in /apps/root will be updated to version 5.34.01.


As described in the three previous emails about the software upgrade,
there are  five significant changes to the systems:

1. New User Certificate System.

The current user certificate will not be valid after the upgrade. If you 
have not obtained a new certificate, please acquire a new one using the 
command /site/bin/jcert -create from any CUE supported machine.

2. New Disk Cache and Volatile Management Systems.

The new cache and volatile systems are based on a Lustre global file 
system, which provides better performance and better scalability.

3. Upgraded Jasmine System.

The modified system has several significant improvement on performance, 
scalability and error handling. The user tools have been re-written to 
deliver more useful information during operation as well as on system 
error.  The output of jput and jget will change significantly.

4. Upgraded Auger System.

The Auger system has been re-engineered to deliver better utilization of 
the farm nodes, cache disks and the tape library system. All Auger 
commands retain the current syntax. However, the output of a new Auger 
command will be in XML format.

5. New Scientific Computing Web Portal.

Finally and most importantly, all user submitted jobs that are queued 
inside the current Auger system at the time of this upgrade need to be 
re-submitted. The reason is incompatibility between old Auger client and 
the new Auger server. We will contact users who have queued jobs at that 
time.

For detailed information about this upgrade, please visit 
https://wiki.jlab.org/cc/external/wiki/index.php/Scientific_Computing#Upcoming_Software_Upgrade

Contacts: Jie.Chen at jlab.org, Sandy.Philpott at jlab.org.



More information about the Jlab-scicomp-briefs mailing list