Swiss Grid Operations Meeting on 2022-01-13 at 14:00

• Zoom link:
  • https://ethz.zoom.us/j/67184143478

*Next meeting: 10 Feb. at 14h00

Minutes:


Derek:

GRAFANA: double check jitter in monitoring plots

Some sites have transfer (rucio) issues with CSCS - being investigated - possible ongoing issue with FTS transfer on kubernetees -provided dCache pools, working on it - see Miguel slides

~80% CPU efficiency for CMS jobs

Steven:

limiting the #jobs hitting the scratch increases the efficiency. As expected we’re hit on the overall performance. Next increase the #CPU

Unige-BAOBAB work in progress Q1 2022 to increase the 10Gb/sec

From 17.12 very stable running ! see Miguel's slides for more details

Roland:

Now running straight from Singularity seems to work fine

Miguel CSCS:

Daint: replaced Dell SSD based storage pool with HUAWEI HDD , bigger capacity. This contributed to the stability seen in ATLAS

Current job limits: ATLAS 5000 CPUs CMS 6500 LHCb 3300

Now we can try to increase the #CPU (slightly above pledges to have some margin) ATLAS 7000 CPUs (in two steps of 1000 and check stability ) CMS 7000 LHCb 3300

Mont Fort: no significant progress during the xmas break Alps infrastructures under maintanance

Followup from previous Action Items

Action items

ATLAS

CMS

LHCb

T2 Sites reports

• ATLAST2Report-Dec2021.pdf: ATLAS CH Tier2 report

CSCS

• RRB values for ATLAS-CSCS CPU in 2021: 74.24k HS06
• RRB values for ATLAS-CSCS Disk in 2021: 2574 TB

UNIBE

• LHEP working on estension (compute and scratch file system)
• UBELIX stable

T3 Sites reports

PSI

UniGe

• BAOBAB full integration with ATLAS still problematic, working on it, needs network upgrade

EGI / WLCG

N.T.R.

Review of open tickets

• Attendants
• CSCS: Dino, Marco Induni, Miguel, Dario, Elia, Gianni
• CMS: Derek, Clemens, Mauro
• ATLAS: Steven,
• LHCb: Lesya
• EGI: