Data Acquisition System Proposal for the PANDA Experiment

PANDA is a new experiment at the FAIR facility Germany that will start in 2020. The experiment will produce enormous quantity of data. The construction of data acquisition system poses an architectural and implementation challenge. The rich physics program of the experiment does not allow for construction of hardware trigger systems. The trigger-less approach starts with event building which will be followed by a software filtering algorithms to select interesting events. The processing time may vary depending on contents of the data and intelligent assignment of the processing farm for subsequent events will influence latency of the system and maximal rate. In this paper, we will discuss possible scenarios of load balancing mapped onto possible architecture of the Panda DAQ system . We will also discuss architectural issues of the data acquisition and compare the requirements against those of established experiments at CERN. The proposal will be accompanied by simulation data showing the performance of the proposed solution.