Review of action items from Tier2 meeting at SLAC: NotesTier2Nov30. Overarching near term goals (December 15) are:
Establish 200 MB/s sustained throughput to all Tier2s
Establish analysis queues at all Tier2s
Replicate Rel 12 AODs to all Tier2, for routine pathena analysis
Operations: Production (Kaushik)
Production summary (Kaushik)
All going fine. BU solved issue w/ storage endpoint. AGLT2 - during site reports.
Follow-up on known leaks in pileup jobs - stopped submitting those jobs.
Production shift report (Nurcan/Mark)
Wensheng - OU-OSCER: problem w/ copying output to storage - Paul investigating. UTD NFS problems.
There is a task that is generating very large files > 2 GB. These jobs are failing at BNL, but not at UC_ATLAS_MWT2. Need to take this issue w/ Paul.
eLog December 15 - status (Mark) - will start using it today to try out. Next week will make this available official.
Follow-up on ADC Operations plan to submit to Alexei. Kaushik will send to ATLAS management today. Note January 21-22 at CERN there will be a combined shift training meeting.
Operations: DDM (Alexei)
DQ2 0.5.0 schedule and plan (Hiro)
Will install at BNLDISK and BNLTAPE (not BNLPANDA which will impact production). Today/tomorrow.
Patrick will take a look today.
Charles will also take a look.
Kaushik reports that France and UK have seen improvements using OSG 0.5. Backlogs clearing much more quickly.
Follow-up on LRC upgrade project
John has LFC running, and is doing internal testing. Will load up and check performance. Learning client tools - eg. how to add additional tools.
John will convene an LFC working group.
Priority is to bring up an instance of a LFC
Milestone of December 20 - public LFC at BNL, ready for clients to register
Follow up on AOD replication for analysis at Tier2s - will resume at all sites. - Status
We looked at AMANDA status - generated more questions than answers.
Follow-up Four sites are various states of implementation:
SLAC - there may be an issue with the pilot copying files out - Paul looking into this.
MWT2 - Charles has setup the Condor config. Defining siteinfo config. Mark will send test pilots.
OU - May still be an issue: will submit some test jobs
BU - ready for test jobs.
SWT2_UTA - done * Can we agree that we have this mileston completed by December 15? Yes. * Follow-up Can we run jobs on a regular basis, and collect information? Mark will automate submission of pathena test jobs.
SWT2_UTA still being addressed. Need VORS registration.
BNL accounting info was lost - Xin investigating. There was confusion on the WLCG APEL site - having to do with the change in the Gratia site name - they appear to have static mappings. Xin still investigating.
Throughput initiative - overview (Shawn)
See notes from dedicated meeting this week: MinutesTPDec10
Need to document storage endpoints, and to benchmark the storage w/ either bonnie++ and iozone.
Dantong: looking into increasing memory on gridftp doors, but can't - these are old systems.
Wei notes that SLAC sees CPU limiting.
Will meet again on Monday.
OSG
OSG site administrators meeting at Fermilab: Dec 12-13
Please note that this site is a content mirror of the BNL US ATLAS TWiki. To edit the content of this page, click the Edit this page button at the top of the page and log in with your US ATLAS computing account name and password.