OSG Area Coordinator’s Report: Workload Management May14 th, 2009 Maxim Potekhin BNL

2 Overview Workload Management Current Initiatives: extensions of the Panda job aggregation and submission service, in particular data movement and security enhancements  WLCG interoperability: establishing Panda Pilot submission to EGEE sites where SCAS has been deployed  Continued coordination with our EGEE colleagues  SCAS deployment close to finish in France (IN2P3), UK Cloud (Lancaster), and in Spain (IFIC) – basic deployment done but details need tweaking  Non-Atlas VOs:  Scheduling next round of CHARMM calculations (using sites other than Purdue)  Generic Panda Data Movement Accomplishments Since Last Report  Presentation about Panda at North Carolina Grid School  Panda Pilot for Atlas: the Pilot code modified, to ensure operation of the existing automated data movement facility, when user identity is changed via setuid (gLExec)  Experience with integration of glexec into the pilot-based framework documented in our CHEP paper (already helpful for our EGEE colleagues)  Generic Panda Data Movement prototype created:  Leveraging the Pilot framework to simplify data delivery for non-Atlas VOs

3 Overview Workload Management Issues / Concerns  Finishing work on the Generic Panda Data Mover in a timely fashion is important for furthering our Engagement effort in the part that includes Panda  Wrapping up Panda Pilot code for Atlas (is now close)  Both Pilot work and SCAS deployment face an aggressive schedule