A. Kumary, Partha Kunduz, A.P. Singhx, Li-Shiuan Pehy, N. K. Jhay
{"title":"A 4.6Tbits/s 3.6GHz single-cycle NoC router with a novel switch allocator in 65nm CMOS","authors":"A. Kumary, Partha Kunduz, A.P. Singhx, Li-Shiuan Pehy, N. K. Jhay","doi":"10.1109/ICCD.2007.4601881","DOIUrl":null,"url":null,"abstract":"As chip multiprocessors (CMPs) become the only viable way to scale up and utilize the abundant transistors made available in current microprocessors, the design of on-chip networks is becoming critically important. These networks face unique design constraints and are required to provide extremely fast and high bandwidth communication, yet meet tight power and area budgets. In this paper, we present a detailed design of our on-chip network router targeted at a 36-core shared-memory CMP system in 65 nm technology. Our design targets an aggressive clock frequency of 3.6 GHz, thus posing tough design challenges that led to several unique circuit and microarchitectural innovations and design choices, including a novel high throughput and low latency switch allocation mechanism, a non-speculative single-cycle router pipeline which uses advanced bundles to remove control setup overhead, a low-complexity virtual channel allocator and a dynamically-managed shared buffer design which uses prefetching to minimize critical path delay. Our router takes up 1.19 mm2 area and expends 551 mW power at 10% activity, delivering a single-cycle no-load latency at 3.6 GHz clock frequency while achieving apeak switching data rate in excess of 4.6 Tbits/sper router node.","PeriodicalId":6306,"journal":{"name":"2007 25th International Conference on Computer Design","volume":"163 1","pages":"63-70"},"PeriodicalIF":0.0000,"publicationDate":"2007-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"212","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"2007 25th International Conference on Computer Design","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/ICCD.2007.4601881","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 212
Abstract
As chip multiprocessors (CMPs) become the only viable way to scale up and utilize the abundant transistors made available in current microprocessors, the design of on-chip networks is becoming critically important. These networks face unique design constraints and are required to provide extremely fast and high bandwidth communication, yet meet tight power and area budgets. In this paper, we present a detailed design of our on-chip network router targeted at a 36-core shared-memory CMP system in 65 nm technology. Our design targets an aggressive clock frequency of 3.6 GHz, thus posing tough design challenges that led to several unique circuit and microarchitectural innovations and design choices, including a novel high throughput and low latency switch allocation mechanism, a non-speculative single-cycle router pipeline which uses advanced bundles to remove control setup overhead, a low-complexity virtual channel allocator and a dynamically-managed shared buffer design which uses prefetching to minimize critical path delay. Our router takes up 1.19 mm2 area and expends 551 mW power at 10% activity, delivering a single-cycle no-load latency at 3.6 GHz clock frequency while achieving apeak switching data rate in excess of 4.6 Tbits/sper router node.