Hongwei Kan, Rui Hao, Jiangyao Wang, G. Mei, Dongdong Su, Songqing Deng
{"title":"基于HLS的超低延迟FAST协议解码器","authors":"Hongwei Kan, Rui Hao, Jiangyao Wang, G. Mei, Dongdong Su, Songqing Deng","doi":"10.1145/3487075.3487150","DOIUrl":null,"url":null,"abstract":"Nowadays the advantages of heterogeneous acceleration technology are becoming more and more obvious. Edge computing center began to accelerate its distributed trading business based on FPGA technology, especially in the field of securities. Shanghai stock market uses efficient FIX Adapted for Streaming (FAST) protocol to transmit the market data. FAST protocol has high compression ratio and complex decoding process. Aiming at efficient FAST protocol decoding, we propose a new general FPGA logic design framework, which implements a low complexity parallel FAST field matching state machine and a low latency stock code mapping mechanism. In this paper, Vivado HLS is used to design and generate the relevant codes. Test results show that the clock frequency after place and routing can reach 250MHz under Xilinx VU37P FPGA. For the 10-level and 50-order message of FAST decoding, the processing delay of core parsing module is only 36ns and 56ns, FPGA subsystem delay is about 120ns. The solution in the paper is one order of magnitude more than traditional CPU software solutions, and is superior to other existing designs base on FPGA, so it is expected to be widely used in the financial market.","PeriodicalId":354966,"journal":{"name":"Proceedings of the 5th International Conference on Computer Science and Application Engineering","volume":"32 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2021-10-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"HLS Based Ultra-low Latency FAST Protocol Decoder\",\"authors\":\"Hongwei Kan, Rui Hao, Jiangyao Wang, G. Mei, Dongdong Su, Songqing Deng\",\"doi\":\"10.1145/3487075.3487150\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"Nowadays the advantages of heterogeneous acceleration technology are becoming more and more obvious. Edge computing center began to accelerate its distributed trading business based on FPGA technology, especially in the field of securities. Shanghai stock market uses efficient FIX Adapted for Streaming (FAST) protocol to transmit the market data. FAST protocol has high compression ratio and complex decoding process. Aiming at efficient FAST protocol decoding, we propose a new general FPGA logic design framework, which implements a low complexity parallel FAST field matching state machine and a low latency stock code mapping mechanism. In this paper, Vivado HLS is used to design and generate the relevant codes. Test results show that the clock frequency after place and routing can reach 250MHz under Xilinx VU37P FPGA. For the 10-level and 50-order message of FAST decoding, the processing delay of core parsing module is only 36ns and 56ns, FPGA subsystem delay is about 120ns. The solution in the paper is one order of magnitude more than traditional CPU software solutions, and is superior to other existing designs base on FPGA, so it is expected to be widely used in the financial market.\",\"PeriodicalId\":354966,\"journal\":{\"name\":\"Proceedings of the 5th International Conference on Computer Science and Application Engineering\",\"volume\":\"32 1\",\"pages\":\"0\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2021-10-19\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Proceedings of the 5th International Conference on Computer Science and Application Engineering\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.1145/3487075.3487150\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Proceedings of the 5th International Conference on Computer Science and Application Engineering","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1145/3487075.3487150","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 0
摘要
目前,异构加速技术的优势越来越明显。边缘计算中心开始加速其基于FPGA技术的分布式交易业务,特别是在证券领域。上海证券市场采用高效的FIX for Streaming (FAST)协议传输市场数据。FAST协议压缩比高,解码过程复杂。针对高效的FAST协议解码,提出了一种新的通用FPGA逻辑设计框架,实现了低复杂度并行FAST字段匹配状态机和低延迟库存码映射机制。本文使用Vivado HLS来设计和生成相关的代码。测试结果表明,在Xilinx VU37P FPGA下,放置和布线后的时钟频率可达到250MHz。对于FAST解码的10级和50阶消息,核心解析模块的处理延迟仅为36ns和56ns, FPGA子系统的处理延迟约为120ns。本文的解决方案比传统的CPU软件解决方案高出一个数量级,并且优于现有的基于FPGA的其他设计,因此有望在金融市场上得到广泛应用。
Nowadays the advantages of heterogeneous acceleration technology are becoming more and more obvious. Edge computing center began to accelerate its distributed trading business based on FPGA technology, especially in the field of securities. Shanghai stock market uses efficient FIX Adapted for Streaming (FAST) protocol to transmit the market data. FAST protocol has high compression ratio and complex decoding process. Aiming at efficient FAST protocol decoding, we propose a new general FPGA logic design framework, which implements a low complexity parallel FAST field matching state machine and a low latency stock code mapping mechanism. In this paper, Vivado HLS is used to design and generate the relevant codes. Test results show that the clock frequency after place and routing can reach 250MHz under Xilinx VU37P FPGA. For the 10-level and 50-order message of FAST decoding, the processing delay of core parsing module is only 36ns and 56ns, FPGA subsystem delay is about 120ns. The solution in the paper is one order of magnitude more than traditional CPU software solutions, and is superior to other existing designs base on FPGA, so it is expected to be widely used in the financial market.