{"title":"探索具有单周期表达路径和自学习节流的高性能无缓冲noc","authors":"Bhavya K. Daya, L. Peh, A. Chandrakasan","doi":"10.1145/2897937.2898075","DOIUrl":null,"url":null,"abstract":"Router buffers are the main reason for the Network-on-Chip's (NoC) scalable bandwidth, but consumes significant area and power. The SCEPTER bufferless NoC sets up single-cycle virtual express paths dynamically, allowing packets to traverse non-minimal paths without latency penalty. Using prioritization, bypassing, and throttling mechanisms, we maximize opportunities to use these paths while pushing bandwidth. For 64 and 256 nodes, we achieve 62% lower latency, 1.3× higher throughput, and 35% lower starvation over a baseline bufferless NoC for synthetic traffic. Full-system 36-core simulations show a 19% lower runtime, on-par performance to a buffered network, with 36% lower area, 33% lower power.","PeriodicalId":185271,"journal":{"name":"2016 53nd ACM/EDAC/IEEE Design Automation Conference (DAC)","volume":"15 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2016-06-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"21","resultStr":"{\"title\":\"Quest for high-performance bufferless NoCs with single-cycle express paths and self-learning throttling\",\"authors\":\"Bhavya K. Daya, L. Peh, A. Chandrakasan\",\"doi\":\"10.1145/2897937.2898075\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"Router buffers are the main reason for the Network-on-Chip's (NoC) scalable bandwidth, but consumes significant area and power. The SCEPTER bufferless NoC sets up single-cycle virtual express paths dynamically, allowing packets to traverse non-minimal paths without latency penalty. Using prioritization, bypassing, and throttling mechanisms, we maximize opportunities to use these paths while pushing bandwidth. For 64 and 256 nodes, we achieve 62% lower latency, 1.3× higher throughput, and 35% lower starvation over a baseline bufferless NoC for synthetic traffic. Full-system 36-core simulations show a 19% lower runtime, on-par performance to a buffered network, with 36% lower area, 33% lower power.\",\"PeriodicalId\":185271,\"journal\":{\"name\":\"2016 53nd ACM/EDAC/IEEE Design Automation Conference (DAC)\",\"volume\":\"15 1\",\"pages\":\"0\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2016-06-05\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"21\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"2016 53nd ACM/EDAC/IEEE Design Automation Conference (DAC)\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.1145/2897937.2898075\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"2016 53nd ACM/EDAC/IEEE Design Automation Conference (DAC)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1145/2897937.2898075","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
Quest for high-performance bufferless NoCs with single-cycle express paths and self-learning throttling
Router buffers are the main reason for the Network-on-Chip's (NoC) scalable bandwidth, but consumes significant area and power. The SCEPTER bufferless NoC sets up single-cycle virtual express paths dynamically, allowing packets to traverse non-minimal paths without latency penalty. Using prioritization, bypassing, and throttling mechanisms, we maximize opportunities to use these paths while pushing bandwidth. For 64 and 256 nodes, we achieve 62% lower latency, 1.3× higher throughput, and 35% lower starvation over a baseline bufferless NoC for synthetic traffic. Full-system 36-core simulations show a 19% lower runtime, on-par performance to a buffered network, with 36% lower area, 33% lower power.