Supercomputer Time: Theory & Its Dependencies

[Noxide]

I've heard scarce references to supercomputer time with respect to some of the work theoretical physicists do. Is there a branch of theory that is absolutely dependent on supercomputer time? Are there any branches of theory that do not require supercomputer time?

 

[diazona]

You do know what a supercomputer is, right?

Physicists just use them whenever they have exceptionally complicated computations to do, or large simulations to run. Basically whenever you have a computer program to run, you find the computer that is the best match for the requirements of the program, and if it's a particularly demanding program, you might need a supercomputer.

Pretty much all branches of physics use computers these days, for all sorts of calculations.

 

[richardbruce]

whether modeling, earthquake simulation and nuclear explosion / reactor simulations are the big examples I know of for running on a super computer , but I'm sure there are many more.

A super computer is basically just a lot of computers tightly networked together and in fact people occasionally build there own out of commodity PC's and routers. This means that for your physics to work well on a super computer you need to write a very highly threaded algorithm so that lots of things can happen in parallel.The big name in super computing is Cray, but even they are kept from going under by America defense spending (machine in the CIA basement cracking AES) :).