I'd like to set up a cluster for processing biological data.
I've seen some of the possible solutions that are around, but its hard to see what difference any of them make. Some people say to use MOSIX, some reconmend using batch systems such as OpenPBS or Sun Grid Engine.
What to choose? How to go about this?
I personally feel most comfortable with Debian Linux, are these tools available for Debian ?
Comments
timely query
Some of Team Nodal have been wrestling with these issues on and off for some weeks now. Expect the details soon.
As I recall, we looked at all the options you mention, started out with OpenPBS, then decided that Sun Grid Engine was worth a look.
I'm quite intrigued by Jim Kent's Parasol too, if only because the config details look nice and short.
For sure, anything you want works under Debian/Linux.