Re: distributed batch processing
On May 11, 2005, at 4:03 PM, Drake Diedrich wrote:
Sun's Grid Engine (http://gridengine.sunsource.net)
which is a reincarnation of Codine, which was a reincarnation
DQS/SGE/PBS/*NQS are all implementations of the standard
POSIX queueing systems.
Actually, I was trying to make a different point: DQS was
as Codine. The company that commercialized it extended it to
Sun purchased the company and renamed GRD GridEngine.
You said it better than I did:
SGE isn't just a reincarnation, it's clearly a fork of the same
Which is what I was intending to convey. DQS, GRD, and GridEngine
share a common code lineage. Same with Condor and LoadLeveler,
though I think the latter has diverged wildly from the former by now.
Debian really ought to have at least one packaged
implementation of this (DQS was, but I couldn't keep it working).
As Josh mentioned, the Torque fork of OpenPBS is probably the
best bet. Torque is OpenPBS + FOSS extensions, including
re-implementation of PBSPro features. There are already some
debs available and there's at least one script floating around
to build deb packages.
It's probably also worth looking at the maui scheduler (which
runs on top of torque or other batch systems). Someone wiser
than I should probably browse the licenses at
www.clusterresources.com to verify that torque and/or maui
meet the Debian standards before any work is started.
/* Dale Southard Jr. email@example.com 925-422-1463 f:422-9429
* Computer Scientist, Advanced Simulation and Computing Program
* Lawrence Livermore National Lab, L-073, Livermore, CA 94551