So… you’ve heard me go on about this computer cluster, but I’ve still got a long ways to go before things are up and running. I’ll try to keep track of all my progress and post it in the category called Cluster.
In the end, we decided to install CentOS on our systems, since it’s been recommended to us by many, and is based off Red Hat Enterprise Linux (RHEL).
We’re probably going to be using TORQUE as our resource manager, and Maui or Moab as our Scheduler. Moab comes with a lot of features, so even though it’s a commercial licence, it’s a cheap commercial licence for what it offers. It’s also what’s used on WestGrid, which is a big name around UBC if you’ve got some number crunching to do.
To make future administration a bit easier, we’re hoping to have a single master system, and then the rest of the nodes be network booted slave systems. That’s what I’m hoping to learn first, since that’s really the major configuration that’s going to happen. After that, everything is going to be handled by the resource manager and scheduler…
Keep looking for more info on this as I learn more… 🙂