> Various reasons are given, but I think the biggest one is that many
> sites (and individual WNs) are misconfigured or broken and the failure
> rate for jobs submitted in the conventional way is too high. Pilot jobs
> can check their environment before pulling the real job so the effective
> failure rate is much lower.
This doesn't need the multi-user scheme they are using which is the
reason they require glexec. The reason they want the multi-user/glexec
scheme is to do a last second intra-VO job scheduling.
cheers
alessandra
Burke, S (Stephen) wrote:
> Testbed Support for GridPP member institutes
>> [mailto:[log in to unmask]] On Behalf Of Kostas Georgiou said:
>> I could spend hours explaining why glexec is a bad idea (and I will),
>> but before that I would like someone to tell me why we need the
>> pilot jobs in the first place.
>
>
> Stephen
>
--
***********************************
* Alessandra Forti *
* NorthGrid Technical Coordinator *
* University of Manchester *
***********************************
|