[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: Debian Science related blog post (about Debian usage @ EDF)



On Thu, 27 May 2010, Jerome Robert wrote:

Hi all,

Hi Alexandre, it's been a long time !

We are also interested in using Debian on our cluster. We are currently using SystemImager on Redhat for the deployment. Our file server nodes have already been migrated to Squeeze. I'm still thinking about how to migrated computation nodes. http://www.debianclusters.org is a great help but it looks a bit dormant now.

I'm wondering what scheduler and resource manager we should use:
- Grid Engine which has been in Debian for years
- Torque which has recently entered in Sid (but without Maui)

You could also take a look at slurm (slurm-llnl in debian), which is rather popular for new installs. New versions also have most of the scheduling logic that you earlier needed Maui for, but I don't know how far those versions have made it in debian.

Concerning deployment, I saw that SystemImager debian package has been removed and that the upstream developments have almost stopped 2 years ago. I had a look to FAI which seems to be a very active project.

I have good experience with FAI too.

/Mattias Wadenstein


Reply to: