Super User is a question and answer site for computer enthusiasts and power users. Join them; it only takes a minute:

Sign up
Here's how it works:
  1. Anybody can ask a question
  2. Anybody can answer
  3. The best answers are voted up and rise to the top

I want to install a Grid Engine cluster entirely on my laptop, the "cluster" having only two nodes, one head node and one node for computing. The nodes should run within the same instance of the same OS (ubuntu 11.10), and each "node" should be assigned to one core (I have a laptop with dual-core). My question(s) is: Is this possible, and if yes, how?

Some context on why I want to do that: I am using grid engine at work on a real cluster with many nodes, each node having multiple CPUs, but there I am only allowed to run certain mission-critical computations. I want to try out a few things with array jobs, which are absolutely not intensive to compute, so I just need a grid engine installation to play around. So I was wondering if there is any way to have a grid enine cluster at home, no matter how weak and small; what's important here is whether conceptually it works on the small cluster or not; I do not want to conduct the actual computations on a grid engine cluster on my laptop, I am aware that would be crazy. Rather, I want to play around with my scripts which run the array jobs, and for that I need a cluster which is not the one at work, and which is entirely under my control.

share|improve this question

On Solaris, I would use zones/containers. As you are on Ubuntu, you might try using Linux containers (lxc) or similar OS virtualization.

share|improve this answer

I created a script for doing this without any user interaction on Travis-CI, and it should work on any Ubuntu install. You can check out the gist here.

share|improve this answer

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .