[Swift-user] Using > 1 CPU per compute node under GRAM

[Ben Clifford]

On Mon, 21 Jul 2008, Michael Wilde wrote:

> In the meantime, does anyone know if there's a way to specify
> compute-node-sharing between separate single-cpu jobs via both GRAMs?
> 
> And if this is dependent on the local job manager code or settings? (Ie might
> work on some sites but not others)?

You can specify via GRAM RSL; however at least TGUC deliberately does not 
allow that - one job gets an entire node. I imagine other sites are 
similar.

Coasters should allow this to be done, by running two coaster workers on 
one node. I plan to look at doing that sometime.

> This will ideally be handled as desired by Falkon or Coaster, but in the 
> meantime I was hoping there was a simple setting to give MikeK better 
> CPU yield on Abe.

There isn't.

Him and I have investigated this before, I think. I've just put this in 
the swift bugzilla as bug 150.

--