<!-- keep this as a security measure:
* Set ALLOWTOPICCHANGE =
TWikiAdminGroup,Main.LCGAdminGroup,Main.EgiGroup
* Set ALLOWTOPICRENAME =
TWikiAdminGroup,Main.LCGAdminGroup
#uncomment this if you want the page only be viewable by the internal people
#* Set ALLOWTOPICVIEW =
TWikiAdminGroup,Main.LCGAdminGroup,Main.ChippComputingBoardGroup
-->
Swiss Grid Operations Meeting on 2016-06-06 at 14:00
Site status
CSCS
-
Systems:
- Prepared all Phoenix servers to be shipped out to Gianfranco and Roland
- Installed VM for ARC 6.x tests
- Deployed CentOS 7 Image on Piz Daint
- and re-deployed with the missing unzip package
- Atlas is running on CentOS 7 and also LHCb by mistake (but no one is complaining)
- Storage
Scratch
- No interventions in the last month, stable operation
dCache
- Deployed new nodes
- Using GPFS as backend
- Completed data migration
- (!) CMS pools are 100% full since 3 weeks, still waiting effective action plan from CMS central ops
network usage during dCache migration
PSI
UNIBE-LHEP
- Stable, running unattended
- Setting up ROCKS 7 framework, with CentOS 7 and Slurm 18.08.6, first nodes to be re-installed shortly.
- With a bit of luck, we'll upgrade without downtime (but with a period of reduced caapacity)
- Monthly summary: Pledged: 42k, delivered 23.6k (22k last month)
- Ubelix contributing ~46% (23% typical)
- Running an average >2.2k slots (2.5k typical during the previous pledge period)
WC UNIBE-LHEP:
5month_UNIBE-LHEP:
- Accounting numbers (from scheduler) from last month
Swiss ATLAS statistics
- Hammercloud availability
- ATLAS_HC_last-month:
-
- ANALY_CSCS-HPC: 85% (80% last month)
- CSCS-LCG2-HPC_MCORE: 95% (99% last month)
- ANALY_UNIBE-LHEP: 100% (100% last month)
- ANALY_UNIBE-LHEP-UBELIX: 94% (100% last month)
- UNIBE-* : 100% (99% last month)
- Slots used
- ATLAS_slots:
- CSCS: 4.2k
- UNIBE: 2.2
- CPU consumption
- ATLAS_CPU:
- CSCS 64% (59% last month)
- UNIBE 36% (41% last month)
- Accounting Numbers from the ATLAS dashboard (May 2019) CSCS+UNIBE
- Delivered vs pledged
CSCS-LCG2: pledged 50k, delivered 51k (was 39.4k)
UNIBE-LHEP: pledged 42k, delivered 23.6 (was 22k)
- VO shares at CSCS
- Looking good following the last maintenance
- CSCS_shares-30day:
-
- CSCS_shares-15day:
-
- CSCS_pending-15day:
-
-
UNIBE-ID
- ARC CE is running smoothly
- Increased slot usage limit for ATLAS jobs from 600 to 1200
- after having observed that the limit of 600 slots is not per partition but global
- more increase possible, e.g all => 600, atlas-preempt => 1200 to make better use of idle resources
- but often only few jobs got sent to atlas-preempt partition though a lot of resources would had been free (weekends)
- tbd with Gianfranco
UNIGE
- No news about the ARC re-deployment
- Considering the proposal of closing the site in GOCDB
NGI_CH
- Certificates
- Obtained two certificates from QuoVadis to check:
- One personal
- One Server (ARC CE)
- Was able to perform basic operations with them: validated
- Next steps:
- Sign a contract (this week?)
- Trust/Link setup with admin rights to us
- Fine-tuning the templates
- ???
- Operations
- Yearly review of GOCDB information
- NGI-CH Open Tickets review
Other topics
Next meeting date:
A.O.B.
Attendants
- CSCS:
- CMS:
- ATLAS:
- LHCb:
- EGI:
Action items
- Item1
-
- dcache migration:
- ATLAS_HC_last-month: