0

I need to perform large amount of computations on a something resembling a cluster, the hardware and the OS are identical (the OS is ubuntu) but no central management software or grid engine is installed. The web search results in mostly outdated or proprietary software.

I hope my question is not too general but, what are the cluster management and job scheduling options for Debian and its derivatives?

For the general management of the cluster I use cssh but this approach is not very efficient when it comes to job scheduling and monitoring. I have experience using the venerable Sun grid engine RIP.

Thanks for reading this!

0

If you're talking about things currently available within the Debian/Ubuntu repositories, I know of at least two options offhand.

  • There is an active free fork of SGE, Son of Gridengine. The packages currently in Debian/Ubuntu are based on this fork. See the gridengine-master and gridengine-exec packages.

  • SLURM. See the slurm-wlm package.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.