[sorry for the terrible triple cross-post, but this topic is in parallel
discussion for the curren LHCb production group]
Stephen Burke wrote:
> The experiments can negotiate to have the queues set up in a way which
> suits them, I don't think there's anything magic about 12 hours. For
> sites which mostly support a single VO they should be prepared to do it
> however you want. Other sites will have to do a bit of arbitration, but
> probably the needs of the experiments aren't so different that you can't
> get a reasonable compromise. Potentially you can have VO-specific queues
> even if the resources they feed are shared.
Excellent. Then I would like to go about requesting that every site
which accepts LHCb VO jobs will provide 24 hours CPU time for the queues
which get "preference". How do I do this?
Actually, now that I think about it, I thought the point of LCG was that
I *wouldn't* have to negotiate with every site these sorts of details.
Suddenly it is seeming like we (LHCb) need to:
1. Install our own software.
2. Negotiate our own queue configurations.
3. Run our own RB and configure which sites we can access.
Given all of that, I think life would be easiest if we could just
gsi-ssh to the CE and just "do our own thing" there with software
installation and job submission.
Cheers,
Ian.
--
Ian Stokes-Rees [log in to unmask]
Particle Physics, Oxford http://www-pnp.physics.ox.ac.uk/~stokes
|