{"title":"一种非易失性主存存储系统","authors":"Michael Wu, W. Zwaenepoel","doi":"10.1145/195473.195506","DOIUrl":null,"url":null,"abstract":"We are designing a machine with a very large amount of nonvolatile solid state memory. While some mechanical storage device may remain present to handle overflow and to protect against catastrophic (software) failures, the nonvolatile solid state memory is the primary repository of permanent data. Benefits of this approach include much improved access times, and, hopefully, simplified storage system software. We argue that technological advances will soon make it possible to build high-end file servers and database servers with large nonvolatile solid state memories, at a price/performance ratio competitive with disk- or RAID-based designs. The eNVy project comprises two largely separate aspects. In terms of software, we are investigating how to write a storage server that can take advantage of a large nonvolatile memory, regardless of what technology is used. In terms of hardware, we are looking at different ways of building such a memory system.<<ETX>>","PeriodicalId":345070,"journal":{"name":"Proceedings of IEEE 4th Workshop on Workstation Operating Systems. WWOS-III","volume":"264 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"1993-10-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"437","resultStr":"{\"title\":\"eNVy: a NonVolatile main memory storage system\",\"authors\":\"Michael Wu, W. Zwaenepoel\",\"doi\":\"10.1145/195473.195506\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"We are designing a machine with a very large amount of nonvolatile solid state memory. While some mechanical storage device may remain present to handle overflow and to protect against catastrophic (software) failures, the nonvolatile solid state memory is the primary repository of permanent data. Benefits of this approach include much improved access times, and, hopefully, simplified storage system software. We argue that technological advances will soon make it possible to build high-end file servers and database servers with large nonvolatile solid state memories, at a price/performance ratio competitive with disk- or RAID-based designs. The eNVy project comprises two largely separate aspects. In terms of software, we are investigating how to write a storage server that can take advantage of a large nonvolatile memory, regardless of what technology is used. In terms of hardware, we are looking at different ways of building such a memory system.<<ETX>>\",\"PeriodicalId\":345070,\"journal\":{\"name\":\"Proceedings of IEEE 4th Workshop on Workstation Operating Systems. WWOS-III\",\"volume\":\"264 1\",\"pages\":\"0\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"1993-10-14\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"437\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Proceedings of IEEE 4th Workshop on Workstation Operating Systems. WWOS-III\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.1145/195473.195506\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Proceedings of IEEE 4th Workshop on Workstation Operating Systems. WWOS-III","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1145/195473.195506","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
We are designing a machine with a very large amount of nonvolatile solid state memory. While some mechanical storage device may remain present to handle overflow and to protect against catastrophic (software) failures, the nonvolatile solid state memory is the primary repository of permanent data. Benefits of this approach include much improved access times, and, hopefully, simplified storage system software. We argue that technological advances will soon make it possible to build high-end file servers and database servers with large nonvolatile solid state memories, at a price/performance ratio competitive with disk- or RAID-based designs. The eNVy project comprises two largely separate aspects. In terms of software, we are investigating how to write a storage server that can take advantage of a large nonvolatile memory, regardless of what technology is used. In terms of hardware, we are looking at different ways of building such a memory system.<>