19.01.26
Present:
David Britton , David Colling, Davide Costanzo, Alastair Dewhurstn Tony Doyle, Katy Ellis, Alessandra Forti, Peter Gronbech, Jonathan Hays, Roger Jones, David Kelsey, Steve Lloyd, Andrew McNab, Andrew Sansum, Sam Skipsey & Jill Sambrook (Minutes)
Apologies:
Daniela Bauer & Peter Clarke
0) Tier-1 Procurement Update [AD]
Email sent by AD prior to the meeting with full details of final set of procurements (attached to agenda)
Summary discussed at the meeting.
Lead times should be ok, and delivery should arrive in good time. Useful features allow us to test future networking requirements, amongst others, and provide some GPU resource.
1) LHCONE Costs? [DB]
Lancaster were slightly surprised that JISC wanted them to pay a fee for LHCONE (unknown fee at this point). Not unreasonable for there to be a charge for connection, but we will need to find out more. Matt D is investigating – initial offer was for additional capacity as well (which is not needed). RJ will investigate further and see where this goes.
Others at sites with LHCONE investigating their situation.
2)CHEP Attendance [SS]
SS has been informed of 7 people looking for GridPP contribution to present talks. This is the numbers at the moment, plus 3 from the IAC/PCC.
Flights are cheap at the moment compared to previous estimates. Encourage people to book now (and charge against this year’s travel budget if your institute can)
Policy as always is to fund people who have talks 50%, with some flexibility for others depending on local needs. SS to apply policy.
Abstract deadline was Friday, but abstract acceptance notifications are 17th Feb.
STANDING ITEMS
SI-0 Tier-1 Manager’s weekly report & Technical Meetings [AD]
A few GGUS tickets have come in over the last week. Relatively small things that are currently being looked at.
LHCb ticket on specific issue with specific vector read patterns at high load – Alex making progress with this.
ATLAS – export to Tape issue being looked at now
CMS new ticket being looked at.
Farm full, Echo busy and functioning. Tape system currently in use and busy at the moment as ALICE are moving their previous run to tape from disk.
Previous procurement update also given at the meeting.
SI-1 ATLAS Weekly Review and Plans [DCos]
-Brij followed up on the issue of low jobs over the holiday period. Should have been fixed last week and occupancy went up to 100%.
Another different issue has occurred this weekend, but it looks like that has been sorted. Another week below pledge, but resolved now.
-tier2 – delivering over pledge.
DB noted that ATLAS are not using the ARM resource at Glasgow – DCos noted that Brij has been at the ADC meetings so should be asked to prod this.
-Manchester on scheduled downtime. AF confirmed it’s potentially another 2 weeks of downtime.
SI-2 CMS Weekly Review and Plans [DC, KE]
-GPU ticket – Brij looking into CMS job
-tier 1 generally good.
-AAA service traffic high over the weekend due to number of machines.
-Brunel restarted tests after being pushed from Daniela
-Bristol – downtime scheduled for storage, but seems to be impacting other systems.
-KE working on Tape Read tests.
SI-3 LHCb Weekly Review and Plans [AM]
-tier 1 tickets already understood
-tier 2 Manchester and Bristol downtimes already noted.
SI-4 Operations Meeting Report [SS,PG,PC] –
-No ops last week because of IRIS collaboration meeting.
SI-5 LCG Management Board Report of Issues [DB]
-NTR – takes places tomorrow.
Update next week.
SI-6 External Contexts (eg NGI/EGI) [PC/JH]
NTR
JH discussed DSIT meeting agenda and context.
REVIEW OF ACTIONS
Next Meeting
26.01.26