32 Tb/s InfiniBand network monitoring tool for the Run 3 LHCb real-time DAQ

The Data Acquisition System (DAQ) of the Large Hadron Collider beauty (LHCb) experiment at CERN is being completely upgraded for the Run 3 of the LHC accelerator. The previous hardware trigger is abandoned because it would not benefit from the upcoming higher luminosity of the accelerator. Thanks to the detector geometry, in addition to the relatively small event size; a software-based event selection, with a trigger-less readout system is possible for LHCb. This allows online data selection through real-time processing at bunch crossing-rate (which results in 32 Tb/s bandwidth). The new DAQ requires new monitoring tools for network performance control and network diagnosis. The goal of the project is to develop a real-time monitoring tool which obtains the performance data from the InfiniBand switches of the network and exposes it to a Prometheus server for the subsequent visualisation on the Grafana UI.