 |
 |
I've put the usual notes in a convenient/usual location: http://storage.esc.rl.ac.uk/weekly/20141126-minutes.txt In other notes from today - while I was chasing updates from non-LHC VOs, I thought I'd check in on ALICE, too, whom we don't hear from very often (also, like MINOS, they are not using many UK T2s). I had an update from Lee which I didn't bring up in the meeting as we were running out of time. Also on the subject of ALICE, it is "ALICE physics week" this week, so a timely update (or maybe it should have been next week): Lee says: As you probably know ALICE is only using T2 facilities at Birmingham and Oxford. Both these sites work well for us. Birmingham is also providing disk storage. It is some time since I checked but I believe that in total ALICE exceeds its CPU pledge, probably due to the opportunistic use of RAL T1. People in ALICE have told me that it doesn’t matter whether separately the T2s meet the pledge for T2 CPU resources. Storage is more critical for ALICE I believe. There is one slightly unusual aspect of the ALICE T2s which is that ALICE does not have group at Oxford. When these things were decided Oxford was recruited to support Birmingham (which was not able to meet the pledge alone I think) and Birmingham was the only ALICE member institute. Now there is also Liverpool and STFC Daresbury as full members (RAL is even an associate member due to the inner tracker project). We have not pushed for more T2 sites as the pledge is met (it is a while since I checked the details). In principle there would be scope for opportunistic use of other T2 resources (perhaps more so in the period just gone than the coming year!) but this was always difficult due to the requirement for an ALICE VO box. I think the only way forward would be if the ALICE VO box could be deployed via some simple virtual machine image (I don’t know technical how this is just an idea.) -- Scanned by iCritical.