Computing at CERN
Run high level triggers and reconstruction on same cpu farm located at LHCb pit
Send RAW and ESD data over the CDR link (80 Gbps) to computer centre for archive, and first pass analysis
Maintain local storage at pit in case CDR down
- accumulate 2 TB/day, therefore need ᡂ TB
Dispatch AOD and TAG etc to regional centres
During shutdowns, down periods do re-processing of RAW data on the farm in the pit
- read RAW back from computer centre
- send new ESD from pit to computer centre
- full farm available so proceeds at twice the rate
- allows 2-3 reprocessings of complete year’s data
Flexible, efficient and maintainable solution