Management of the LHCb Readout Network

The LHCb event building relies on a readout network based on Gigabit Ethernet (GbE). The readout network is extremely performance critical, sophisticated monitoring at all levels is essential for the successful operation of the data acquisition for the experiment. LHCb uses a commercial supervisory control and data acquisition (SCADA) system called PVSS for its experiment control system (ECS), which consists of varied sub-systems. For the efficient operation of the experiment, it is crucial that all parts of the system are monitored and controlled through the same interface in a coherent way. The network monitoring is implemented in the same framework. We show here how a large scale network can be monitored using tools originally made for industrial supervisory control. The status of the network devices and their ports, their connectivity and the traffic are monitored in this system. Alarms can be issued to inform the experiment operators and online network experts about errors such as dropped packets or broadcast-storms. Reports and long-term monitoring using powerful trending tools are possible.