Description
In this session we will present and discuss how HEP has managed and executed the (by far) largest stress test of any Grid infrastructure(s) to a production level using the largest user community in EGEE: HEP. This has been a very successful exercise and the lessons learned are a very good guide for those communities in EGEE (and beyond - collaborating Grids) needing to stress the Grid infrastructure to a production level. We will share the experiences and procedures of the 4 LHC experiments and also we will provide a general description of the protocols defined in operations, DB and services management, middleware creation and support, to ensure a successful CCRC exercise.
Since CCRC'08 formally ended [post-mortem workshop], many changes have taken place and the experiments have continued their activities at a significant scale. In addition, CERN has now announced the startup schedule for the LHC machine. Finally, planning for data taking and production in 2009 needs to start. All of these issues will be discussed, leading to a roadmap for the short to medium term.
The session will therefore be divided into three topics:
- Summary of the lessons learned from the formal CCRC'08 exercise and from production activities since: requirements for future activities, including operations and application support;
- Immediate needs and short-term goals: the LHC startup, first data taking and (re-)processing of 2008 data;
- Preparation for 2009, including the CCRC'09 planning workshop.
A summary report will be produced and attached to this page.