[torqueusers] lab setup
ycsapo at mines.edu
Thu Aug 23 11:30:47 MDT 2007
-----BEGIN PGP SIGNED MESSAGE-----
Thanks all for the comments - especially Miles, running two moms sounds
like a great idea; and yes, the reasons to have this setup are both your
1) and 2).
To address another question, I wouldn't need to keep the jobs synced
between servers. This setup looks like 24 independent 23-node clusters;
thoroughly defeating the purpose of turning 24 machines into a cluster
but allowing 24 students to experiment with clustering technologies
without stepping too much on each other's toes.
> Hi all,
> I've been tasked with setting up a teaching lab for
> super/clustering/distributed/parallel computing here at the Colorado
> School of Mines. We do have a few computational clusters running torque,
> but this will be a classroom. Instead of designating a machine as the
> head, what I've done is set up every machine as both a head and a node
> of every other machine. Granted, this is not very efficient from a
> computational point of view, but it allows for the greatest flexibility
> for a classroom. Does anybody have any comments about this setup?
torqueusers mailing list
torqueusers at supercluster.org
Academic Computing & Networking
Colorado School of Mines
Green Center Rm 249
Phone: (303) 273-3503
Fax: (303) 273-3475
Email: ycsapo at mines.edu
Please use the following link to open a service request:
With a PC, I always felt limited
by the software available.
On Unix, I am limited only by my knowledge.
- --Peter J. Schoenster
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.6 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org
-----END PGP SIGNATURE-----
More information about the torqueusers