The data-logging system of the trigger and data acquisition for the ATLAS experiment at CERN

The ATLAS experiment is getting ready to observe collisions between protons at a centre of mass energy of 14 TeV. These will be the highest energy collisions in a controlled environment to-date, to be provided by the Large Hadron Collider at CERN by mid 2008. The ATLAS Trigger and Data Acquisition (TDAQ) system selects events online in a three level trigger system in order to keep those events promising to unveil new physics at a budgeted rate of ~200 Hz for an event size of ~1.5 MB. This corresponds to a reduction of O(10
⁵
) from the initial bunch-crossing rate of 40 MHz at nominal operating conditions. This paper focuses on the data-logging system on the TDAQ side, the so-called "Sub-Farm Output" (SFO) system. It takes data from the Event Filter farm, which is the third level trigger, and it streams and indexes the events into different files, according to each event's trigger path. The data files are moved to CASTOR, the central mass storage facility at CERN. The final TDAQ data-logging system has been installed using 6 Linux PCs, holding 24 disks of 500 GB each, managed by three RAID controllers on each PC. The data-writing is managed in a controlled round-robin way among three independent filesystems associated to a distinct set of disks, managed by a distinct RAID controller. This design allows fast I/O, which together with a high speed network permits to minimize the number of SFO nodes. We report here on the functionality and performance requirements on the system, our experience with commissioning it and on the performance achieved,