[GE users] qmaster memory and performance

Beadles, Jeff jeff_beadles at mentorg.com
Mon Apr 11 16:34:54 BST 2005


> From: Raphael Y. Rubin [mailto:rafi at cs.caltech.edu] 
> Sent: Friday, April 08, 2005 12:21 PM
> To: users at gridengine.sunsource.net
> Subject: [GE users] qmaster memory and performance
> 
> We're running sge 6.0u1 and have been seeing issues with the 
> master using
> quite a bit of memory and being a little unresponsive.
> 
> example:
> After submitting ~30000 jobs last night, the master was 
> unresponsive, and
> our shadow took over.  This morning, the master was using over 2.5GB

We do this sort of thing at least a couple of times every day.  What
sort of a system is your master?  We're running on a 4-way Sun V480,
with 8GB.  I can't ever recall seeing the load on the master being >1.5

$SGE_ROOT/utilbin/sol-sparc64/loadcheck typically shows ~6GB memory
free.

We're using local bdb spooling, as well as a NFS shared $SGE_ROOT.

A little more information might help in figuring out what is going on.
Is this a new problem?  Has it ever worked before?  What's the config of
your master?

Regards,
	-Jeff

---------------------------------------------------------------------
To unsubscribe, e-mail: users-unsubscribe at gridengine.sunsource.net
For additional commands, e-mail: users-help at gridengine.sunsource.net




More information about the gridengine-users mailing list