Hadoop Vs. Disco Vs. Condor?

I am trying to find a tool that will manage a bunch of jobs on 100 machines in a cluster (submit the jobs to the machines; make sure that jobs are run etc).

开发者_StackOverflowWhich tool would be more simple to install / manage:

(1) Hadoop?

(2) Disco?

(3) Condor?

Ideally, I am searching for a solution that would be as simple as possible, yet be robust. Python integration is also a plus.

I'm unfamiliar with Disco and Condor, but I can answer regarding Hadoop:

Hadoop pros:

• Robust and proven - probably more than anything else out there. Used by many organizations (including the one I work for) to run clusters of 100s of nodes and more.
• Large ecosystem = support + many subprojects to make life easier (e.g. Pig, Hive)
• Python support should be possible through the streaming MR feature, or maybe Jython?

Hadoop cons:

• Neither simple nor elegant (imho). You'll have to spend time learning.

Have you considered the Sun Grid Engine? http://wikis.sun.com/display/GridEngine/Home .