A Monitoring System for the LHCb Data Flow

The LHCb experiment uses the LHC accelerator for the collisions that produce the physics data necessary for analysis. The data produced by the detector by measuring the results of the collisions at a rate of 40 MHz are read out by a complex data acquisition (DAQ) system, which is summarily described in this paper. Distributed systems of such dimensions rely on monitoring and control systems that account for the numerous faults that can happen throughout the whole operation. With this in mind, a new system was created to extend the monitoring of the readout system, in this case by providing an overview of what is happening in each stage of the DAQ process, starting in the hardware trigger performed right after the detector measurements and ending in the local storage of the experiment. This system, a complement to the current run control (experimental control system), intends to shorten reaction times when a problem occurs by providing the operators with detailed information of where a certain fault is occurring. The architecture of the tool and its utilization by the experiment operators are described in this paper.