engleski

Process Control of an Event Filter Farm for a Particle Physics Experiment Based on Expert System Technology

Modern particle physics experiments observing collisions of particle beams generate large amounts of data. Large trigger and data acquisition systems are built to select on-line the most interesting events and write them to persistent storage. The final stage of this selection process nowadays often happens on large computer farms. The stable and reliable operation of such Event Filter farms is critical for the success of these experiments. In this paper, the current status and plans in developing a Problem Solver based on expert system technology, which could be applied for maintaining reliability and uninterrupted operation of the Event Filter Farm, is described. The proposed Problem Solver has been tested with an Event Filter Farm prototype based on the architecture of the CMS experiment. A performance analysis of the Problem Solver integrated in the existing control system is given.

process control; expert systems; fault detection and diagnosis; recovery

nije evidentirano