US HPC Facilities Meeting

Friday 9 Jun 2017, 11:00 → 12:00 US/Central

Tadashi is preparing Jedi/Harvester for Jumbo Jobs at opportunistic sites and went over slides which will also be shown at the S&C Week in Valencia.

 

Doug is using Stratum-R to transfer in to NERSC/ALCF disk

At NERSC, used up 30M inodes allocated.

The problem with these distrubition techniques is the file duplication

Doug talked to CVMFS:
They agreed to support CVMFS preload + Parrot

Xin:
ATLAS fat container, where is that?
Doug: Wei is producing them for NERSC
but requires 24hrs or more to produce, so you cannot keep up
Also not every HPC supports containers/images
Xin: how big are the images?
Doug: I'll send you via email.

Doug: Data handling discussions
Globus will no longer support GridFTP, globus end-points will not support gridftp & x509
So, we will need to use Globus for HPC centers, we'll need to have some storage that can be Globus + WLCG (rucio). Hiro is setting one up at BNL. Doug will provide code to Rucio to support Globus between Globus end-points and FTS between traditional WLCG end-points.
 

11:00 → 11:15 Harvester

Speaker: Tadashi Maeno (Brookhaven National Laboratory (US))

11:15 → 11:30 OLCF Update

Speakers: Danila Oleynik (Joint Institute for Nuclear Research (RU)), Sergey Panitkin (Brookhaven National Laboratory (US))

11:30 → 11:50 NERSC & ALCF

Speakers: Doug Benjamin (Duke University (US)), Taylor Childers (Argonne National Laboratory (US))