{"title":"利用高带宽光子技术的多 FPGA 网内稳定弧度分类器 [特邀]","authors":"Kenji Mizutani;Yutaka Urino;Takanori Shimizu;Hiroshi Yamaguchi;Shigeru Nakamura;Tatsuya Usuki;Kiyo Ishii;Ryosuke Matsumoto;Takashi Inoue;Shu Namiki;Michihiro Koibuchi","doi":"10.1364/JOCN.530695","DOIUrl":null,"url":null,"abstract":"A modern field-programmable gate array (FPGA) card can be equipped with high-bandwidth memory and high-bandwidth optical interconnection networks. This paper presents an in-network stable radix sorter on an eight-FPGA cluster. Each custom Stratix10 MX2100 FPGA card has up to 819-Gbps memory bandwidth (\n<tex>${51.2}\\;{\\rm Gbps} \\times {16}\\;{\\rm channels}$</tex>\n) and up to 800-Gbps network bandwidth (\n<tex>${25}\\;{\\rm Gbps} \\times {32}\\;{\\rm channels}$</tex>\n) with eight custom embedded optical modules. Our radix sorter for a 32-bit key range consists of eight 4-bit counting sorts optimized to in-network processing. Each counting sort needs only one memory read/write access for improving its throughput. We demonstrated a sorting throughput of 37.2 GB/s and an energy efficiency of 9.2 MB/J for 32-bit key range and 16-GiB data size using eight memory channels with 409.6 Gbps memory bandwidth per FPGA. It can scale up to 256 FPGAs with a sorting throughput of 983 GB/s for a 32-bit key range and 512-GiB data size.","PeriodicalId":50103,"journal":{"name":"Journal of Optical Communications and Networking","volume":"17 1","pages":"A34-A45"},"PeriodicalIF":4.0000,"publicationDate":"2024-11-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"In-network stable radix sorter using many FPGAs with high-bandwidth photonics [Invited]\",\"authors\":\"Kenji Mizutani;Yutaka Urino;Takanori Shimizu;Hiroshi Yamaguchi;Shigeru Nakamura;Tatsuya Usuki;Kiyo Ishii;Ryosuke Matsumoto;Takashi Inoue;Shu Namiki;Michihiro Koibuchi\",\"doi\":\"10.1364/JOCN.530695\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"A modern field-programmable gate array (FPGA) card can be equipped with high-bandwidth memory and high-bandwidth optical interconnection networks. This paper presents an in-network stable radix sorter on an eight-FPGA cluster. Each custom Stratix10 MX2100 FPGA card has up to 819-Gbps memory bandwidth (\\n<tex>${51.2}\\\\;{\\\\rm Gbps} \\\\times {16}\\\\;{\\\\rm channels}$</tex>\\n) and up to 800-Gbps network bandwidth (\\n<tex>${25}\\\\;{\\\\rm Gbps} \\\\times {32}\\\\;{\\\\rm channels}$</tex>\\n) with eight custom embedded optical modules. Our radix sorter for a 32-bit key range consists of eight 4-bit counting sorts optimized to in-network processing. Each counting sort needs only one memory read/write access for improving its throughput. We demonstrated a sorting throughput of 37.2 GB/s and an energy efficiency of 9.2 MB/J for 32-bit key range and 16-GiB data size using eight memory channels with 409.6 Gbps memory bandwidth per FPGA. It can scale up to 256 FPGAs with a sorting throughput of 983 GB/s for a 32-bit key range and 512-GiB data size.\",\"PeriodicalId\":50103,\"journal\":{\"name\":\"Journal of Optical Communications and Networking\",\"volume\":\"17 1\",\"pages\":\"A34-A45\"},\"PeriodicalIF\":4.0000,\"publicationDate\":\"2024-11-13\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Journal of Optical Communications and Networking\",\"FirstCategoryId\":\"94\",\"ListUrlMain\":\"https://ieeexplore.ieee.org/document/10752945/\",\"RegionNum\":2,\"RegionCategory\":\"计算机科学\",\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q1\",\"JCRName\":\"COMPUTER SCIENCE, HARDWARE & ARCHITECTURE\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Journal of Optical Communications and Networking","FirstCategoryId":"94","ListUrlMain":"https://ieeexplore.ieee.org/document/10752945/","RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"COMPUTER SCIENCE, HARDWARE & ARCHITECTURE","Score":null,"Total":0}
In-network stable radix sorter using many FPGAs with high-bandwidth photonics [Invited]
A modern field-programmable gate array (FPGA) card can be equipped with high-bandwidth memory and high-bandwidth optical interconnection networks. This paper presents an in-network stable radix sorter on an eight-FPGA cluster. Each custom Stratix10 MX2100 FPGA card has up to 819-Gbps memory bandwidth (
${51.2}\;{\rm Gbps} \times {16}\;{\rm channels}$
) and up to 800-Gbps network bandwidth (
${25}\;{\rm Gbps} \times {32}\;{\rm channels}$
) with eight custom embedded optical modules. Our radix sorter for a 32-bit key range consists of eight 4-bit counting sorts optimized to in-network processing. Each counting sort needs only one memory read/write access for improving its throughput. We demonstrated a sorting throughput of 37.2 GB/s and an energy efficiency of 9.2 MB/J for 32-bit key range and 16-GiB data size using eight memory channels with 409.6 Gbps memory bandwidth per FPGA. It can scale up to 256 FPGAs with a sorting throughput of 983 GB/s for a 32-bit key range and 512-GiB data size.
期刊介绍:
The scope of the Journal includes advances in the state-of-the-art of optical networking science, technology, and engineering. Both theoretical contributions (including new techniques, concepts, analyses, and economic studies) and practical contributions (including optical networking experiments, prototypes, and new applications) are encouraged. Subareas of interest include the architecture and design of optical networks, optical network survivability and security, software-defined optical networking, elastic optical networks, data and control plane advances, network management related innovation, and optical access networks. Enabling technologies and their applications are suitable topics only if the results are shown to directly impact optical networking beyond simple point-to-point networks.