[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [condor-users] large pools like at UW
- Date: Tue, 18 Nov 2003 08:09:41 -0600
- From: Alain Roy <roy@xxxxxxxxxxx>
- Subject: Re: [condor-users] large pools like at UW
What are the specs on your central manager machine?
I don't know all of the details--Erik Paulson may jump in here. But I do
know that our central manager is a RedHat 7.2 computer, one CPU that is >
2GHz, about 1G of main memory and 2G of swap space.
Also, do you run the usual 5 daemons on that one machine, or do you run
the negotiator elsewhere perhaps?
We run both the collector and the negotiator on one machine. We do not run
a condor_startd on the machine, so we do not allow jobs on the central manager.
What about your checkpoint server, what type of machine is that? How
many checkpoint servers do you have?
Our checkpoint servers seem to change periodically. At the moment, we have
three active checkpoint servers. I think that they each have 100-400G of
space on them. They are set up as RAID, I think, but again I'm not sure
which kind. Maybe Erik can jump in on this one too.
For most of our machines, we randomly select which checkpoint server to use:
CKPT_SERVER_HOST = $RANDOM_CHOICE(choice1, choice2)
Condor Support Information:
To Unsubscribe, send mail to majordomo@xxxxxxxxxxx with
unsubscribe condor-users <your_email_address>