Gordon, JC (John) wrote:
> I can see these were rhetorical questions but I'll answer them anyway.
>
> Experiments want pilot jobs to do 'late binding' of payload to job. i.e.
> wait until a job is actually executing before deciding what it does.
> This is how they chose to manage the relative priority of their work
> within a VO. In the steady state the grid will be full of work with
> queues everywhere so some urgent piece of work may have to wait a very
> long time. This is also a means of delivering global fair shares.
>
To solve the above problem the RB scheduling should be design to handle
Advance Reservation and to priotise job submission by the user
deadlines etc. Thus individual (VOs) don't have to design extra
frameworks.
> A secondary reason for pilot jobs, or maybe the primary one for LHCb, is
> to check out a site before trusting it with a job. The pilot job starts,
> checks out the environment and then gets the real job.
> This gets round the current state that so many sites are badly configured
> or don't advertise their state correctly in the BDII.
Then we should fix the SITES and WN's nodes so that we can provide more
resources for the GRID.
Regards,
Mona
--
Mona Aggarwal- Imperial College
Tel: +442075947809
Email: [log in to unmask]
|