Krishna T. Malladi, Frank A. Nothaft, Karthika Periyathambi, Benjamin C. Lee, C. Kozyrakis, M. Horowitz
{"title":"Towards energy-proportional datacenter memory with mobile DRAM","authors":"Krishna T. Malladi, Frank A. Nothaft, Karthika Periyathambi, Benjamin C. Lee, C. Kozyrakis, M. Horowitz","doi":"10.1145/2366231.2337164","DOIUrl":null,"url":null,"abstract":"To increase datacenter energy efficiency, we need memory systems that keep pace with processor efficiency gains. Currently, servers use DDR3 memory, which is designed for high bandwidth but not for energy proportionality. A system using 20% of the peak DDR3 bandwidth consumes 2.3× the energy per bit compared to the energy consumed by a system with fully utilized memory bandwidth. Nevertheless, many datacenter applications stress memory capacity and latency but not memory bandwidth. In response, we architect server memory systems using mobile DRAM devices, trading peak bandwidth for lower energy consumption per bit and more efficient idle modes. We demonstrate 3-5× lower memory power, better proportionality, and negligible performance penalties for data-center workloads.","PeriodicalId":193578,"journal":{"name":"2012 39th Annual International Symposium on Computer Architecture (ISCA)","volume":"21 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2012-06-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"250","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"2012 39th Annual International Symposium on Computer Architecture (ISCA)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1145/2366231.2337164","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 250
Abstract
To increase datacenter energy efficiency, we need memory systems that keep pace with processor efficiency gains. Currently, servers use DDR3 memory, which is designed for high bandwidth but not for energy proportionality. A system using 20% of the peak DDR3 bandwidth consumes 2.3× the energy per bit compared to the energy consumed by a system with fully utilized memory bandwidth. Nevertheless, many datacenter applications stress memory capacity and latency but not memory bandwidth. In response, we architect server memory systems using mobile DRAM devices, trading peak bandwidth for lower energy consumption per bit and more efficient idle modes. We demonstrate 3-5× lower memory power, better proportionality, and negligible performance penalties for data-center workloads.