用于PANDA流DAQ系统的高性能计算节点的新版本
New version of high performance Compute Node for PANDA Streaming DAQ system
摘要：
PANDA是FAIR/Darmstadt目前正在建设的主要实验之一。它的重点是高强度和高质量的反质子束的物理动量高达15 GeV/c。事件速率高达20 MHz，典型事件大小在1.5 KB和4.5 KB之间。导致高达200 GB/s数据速率。本文介绍了一种基于FPGA和CPU/GPU集群的事件过滤技术的流媒体数据采集系统。计算节点（CN）是该系统中基于FPGA的中央板组件。提出了一种基于ATCA的CN的新版本。它的主要特点包括高速数据传输，大规模数据缓冲能力，以支持复杂决策算法的大延迟，高性能数据处理和以太网连接。第一个测试结果与原型。
英文摘要：
PANDA is one of the major experiments currently under construction at FAIR/Darmstadt. Its focus is physics with high intensity and high quality anti-proton beams with momenta up to 15 GeV/c. Event rates up to 20MHz, and a typical event size between 1.5 KB and 4.5 KB. lead to data rates as high as 200 GB/s. A trigger-less streaming DAQ system is introduced in this paper, featuring event filtering based on FPGAs and a CPU/GPU farm. The Compute Node (CN) is the central board FPGA based component in this system. A new version of the ATCA based CN is presented. Its main features include high speed data transmission, massive data buffering capabilities to support large latency for complex decion algorithms, high performance data processing and ethernet connectivity. First test results with a prototype are presented.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
