Research And Implementation Of The DAQ System For CEE

The CSR External-target Experiment(CEE)will be the first large-scale nuclear physics experimental device that operates in the GeV energy region and is completely independently developed in China.Its establishment is of great value for verifying the relevant theories and microscopic models of nuclear interactions and for studying the properties of quantum chromodynamics(QCD)in low temperature and high density regions.Data Acquisition System(DAQ)is an important part of CEE experiment.It is responsible for quickly and completely collecting interesting events from readout electronics,completing real-time processing tasks such as event building and storing valid events to permanent media such as hard disks.In addition to the original physical data processing,the data acquisition system also has functions such as system configuration and status monitoring.Therefore,a reliable and high-performance data acquisition system is of great significance for the achievement of CEE experimental goals.CEE has many challenges in the design and implementation of its data acquisition system.On the one hand,the expected data rate and effective events rate are as high as 4.5 GB/s and 10 kHz,respectively.At the same time,the CEE experiment also has the possibility of upgrading to a larger detector scale in the future.This places high demands on the processing performance of each module in the data acquisition system and on the scalability of the architecture.On the other hand,CEE has up to 7 types of front-end detector subsystems,as well as functional subsystems such as clock,trigger,and slow control.DAQ needs to establish connections with all the above systems and process all kinds of information flow including raw data,commands,status.This also puts forward higher requirements on the compatibility of its protocol design.The main work of this paper is to design a data acquisition system for CEE experiments that can meet the above challenges and meet various functional and performance requirements.In view of the actual needs of CEE experiments,this paper investigates the DAQ solutions of several large-scale high-energy physics experiments at home and abroad.In-depth research has been carried out on key technologies such as distributed architecture design,protocol design,stream processing module design and performance optimization.Based on the above research,a distributed stream processing data acquisition system with software and hardware coordination is implemented.In terms of architecture,this paper combines two self-developed FPGA generalpurpose readout boards CROB-PXI,CROB-PCIe and server clusters to jointly complete distributed data readout and event building.The advantages and disadvantages of different data interaction models in distributed systems are studied,and a scalable cluster processing scheme suitable for CEE is selected.In terms of protocol design,the paper studies the implementation of different communication protocols in FPGA platform and server platform,designs a data container compatible with various CEE detectors,and unifies the processing methods of various information flows in the whole system.In terms of stream processing module design,this paper implements a series of reusable and freely connected stream processing modules to complete tasks such as operation control,status monitoring,and event building.This paper also studies the influence of different working parameters on the processing performance of key modules on the data stream,and optimizes its design scheme.The test results show that with the existing number of devices,the data acquisition system’s data throughput can reach 9 GB/s,and the event building rate exceeds 20 kHz,which achieves the research goals of this paper and fully meets the requirements of CEE.