{"title":"Scalable Low-Power High-Performance Rack-Scale Optical Network","authors":"Jun Feng, Zhehui Wang, Zhifei Wang, Xuanqi Chen, Shixi Chen, Jiaxu Zhang, Jiang Xu","doi":"10.1109/PHOTONICS49561.2019.00006","DOIUrl":null,"url":null,"abstract":"As large-scale applications are demanding more computation power while Moore’s Law is slowing down, rackscale computing systems are being developed to meet the increasing computation and energy requirements. Optical interconnects have become an alternative to address the performance and energy efficiency issues of rack-scale systems because of their superiority in bandwidth, latency, and power. In this paper, we systematically analyze the rack-scale optical network (RSON) architecture with different path reservation schemes and optical inter-chip networks and the most commonly used architecture for high-performance computing systems. We explore the RSON architecture, floorplan optimized delta optical network (FODON) switch architecture and the preemptive chain feedback (PCF) scheme to optimize multi-domain path reservation. Experimental results show that the RSON with FODON switch and PCF scheme can improve system performance per energy consumption by up to 5x, and around 4x on average, while still maintaining better scalability than state-of-the-art systems.","PeriodicalId":64491,"journal":{"name":"光学与光子学期刊(英文)","volume":"24 1","pages":"1-6"},"PeriodicalIF":0.0000,"publicationDate":"2019-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"2","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"光学与光子学期刊(英文)","FirstCategoryId":"1089","ListUrlMain":"https://doi.org/10.1109/PHOTONICS49561.2019.00006","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 2
Abstract
As large-scale applications are demanding more computation power while Moore’s Law is slowing down, rackscale computing systems are being developed to meet the increasing computation and energy requirements. Optical interconnects have become an alternative to address the performance and energy efficiency issues of rack-scale systems because of their superiority in bandwidth, latency, and power. In this paper, we systematically analyze the rack-scale optical network (RSON) architecture with different path reservation schemes and optical inter-chip networks and the most commonly used architecture for high-performance computing systems. We explore the RSON architecture, floorplan optimized delta optical network (FODON) switch architecture and the preemptive chain feedback (PCF) scheme to optimize multi-domain path reservation. Experimental results show that the RSON with FODON switch and PCF scheme can improve system performance per energy consumption by up to 5x, and around 4x on average, while still maintaining better scalability than state-of-the-art systems.