A high-speed data readout method based on ethernet for particle physics experiment

Abstract

Generally, the readout system for a typical particle physics experiment is implemented with multi electronic modules resided in a standard crate, e.g. VME. Each module receives data from the front-end electronics (FEE) and transmits them to the crate controller for transferring to the data acquisition (DAQ) system in real time. With the increasing physical event rate and number of electronic channels, this readout scheme meets the challenge of improving readout speed caused by the limited bandwidth of crate backplane. One of valid methods is to adopt crates with high performance backplane, e.g. ATCA. In this paper, instead of improving backplane performance, we make each readout module having capability of transmitting data to DAQ. A high-speed data readout method based on Ethernet is designed for each module. Features of explicitly parallel data transmitting and distributed network architecture make the readout system has advantage of adapting varying requirements of particle physics experiments. To guarantee the readout performance and flexibility, a standalone embedded CPU system is utilized for network protocol stack processing. To receive customized data format and protocol from FEE, a field programmable gate array (FPGA) is used for logic reconfiguration. To optimize the interface and improve the data swap speed between CPU and FPGA, a sophisticated method based on SRAM is presented in this paper. In this method, in the aspect of hardware, FPGA fakes itself into a SRAM for CPU. Data from FEE is cached firstly in FPGA and then put on the SRAM bus for CPU fetching. In the aspect of software, Linux operating system pre-allocates physical address for this fake SRAM, and the embedded device driver controls and manages it. To synchronize the data swap, a hand-shaking scheme is put forward in this method. For the purpose of evaluating this high-speed readout method, a simplified readout module is designed and implemented. Test results show that this module can support up to 70Mbps data throughput from the readout module to DAQ smoothly.