完美的页面放置

ACM-SE 35 Pub Date : 1997-04-02 DOI:10.1145/2817460.2817472
Thomas J. Murray, A. A. McRae, A. W. Madison
{"title":"完美的页面放置","authors":"Thomas J. Murray, A. A. McRae, A. W. Madison","doi":"10.1145/2817460.2817472","DOIUrl":null,"url":null,"abstract":"Careful page placement policies are memory management heuristics that map virtual pages to physical memory in a manner that tends to reduce the average CPU cache conflict miss rate compared to random placement. Careful placement is particularly well-suited to large direct-mapped caches. Perfect page placement is the limiting case of careful placement and is achieved by a virtual page mapping that minimizes the conflict miss rate. Cache performance curves for perfect page placement are useful benchmarks for estimating the relative error of other careful placement policies. On-line perfect placement is not realizable because the computation requires future knowledge of the memory reference string. However, off-line perfect placement is a systems modeling tool analogous to the OPT page replacement policy for virtual memory systems; both policies are tools for computing upper bounds on memory system performance. In this paper, perfect page placement is presented and proved NP-hard for demand-paged virtual memory systems with direct-mapped caches and infinite physical memory capacity. A class of perfect placement approximation algorithms is also presented and proved NP-hard.","PeriodicalId":274966,"journal":{"name":"ACM-SE 35","volume":"63 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"1997-04-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"1","resultStr":"{\"title\":\"Perfect page placement\",\"authors\":\"Thomas J. Murray, A. A. McRae, A. W. Madison\",\"doi\":\"10.1145/2817460.2817472\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"Careful page placement policies are memory management heuristics that map virtual pages to physical memory in a manner that tends to reduce the average CPU cache conflict miss rate compared to random placement. Careful placement is particularly well-suited to large direct-mapped caches. Perfect page placement is the limiting case of careful placement and is achieved by a virtual page mapping that minimizes the conflict miss rate. Cache performance curves for perfect page placement are useful benchmarks for estimating the relative error of other careful placement policies. On-line perfect placement is not realizable because the computation requires future knowledge of the memory reference string. However, off-line perfect placement is a systems modeling tool analogous to the OPT page replacement policy for virtual memory systems; both policies are tools for computing upper bounds on memory system performance. In this paper, perfect page placement is presented and proved NP-hard for demand-paged virtual memory systems with direct-mapped caches and infinite physical memory capacity. A class of perfect placement approximation algorithms is also presented and proved NP-hard.\",\"PeriodicalId\":274966,\"journal\":{\"name\":\"ACM-SE 35\",\"volume\":\"63 1\",\"pages\":\"0\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"1997-04-02\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"1\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"ACM-SE 35\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.1145/2817460.2817472\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"ACM-SE 35","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1145/2817460.2817472","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 1

摘要

谨慎的页面放置策略是一种内存管理启发式方法,它将虚拟页面映射到物理内存,与随机放置相比,这种方式倾向于降低平均CPU缓存冲突缺失率。小心放置特别适合大型直接映射缓存。完美的页面放置是仔细放置的极限情况,它是通过虚拟页面映射实现的,该映射可以最小化冲突缺失率。完美页面放置的缓存性能曲线是估计其他谨慎放置策略的相对误差的有用基准。在线完美放置是无法实现的,因为计算需要将来了解内存引用字符串。然而,离线完美放置是一种系统建模工具,类似于虚拟内存系统的OPT页面替换策略;这两种策略都是计算内存系统性能上限的工具。本文提出并证明了具有直接映射缓存和无限物理内存容量的按需分页虚拟内存系统的完美页面放置问题。提出了一类理想的位置逼近算法,并证明了算法的np困难性。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
Perfect page placement
Careful page placement policies are memory management heuristics that map virtual pages to physical memory in a manner that tends to reduce the average CPU cache conflict miss rate compared to random placement. Careful placement is particularly well-suited to large direct-mapped caches. Perfect page placement is the limiting case of careful placement and is achieved by a virtual page mapping that minimizes the conflict miss rate. Cache performance curves for perfect page placement are useful benchmarks for estimating the relative error of other careful placement policies. On-line perfect placement is not realizable because the computation requires future knowledge of the memory reference string. However, off-line perfect placement is a systems modeling tool analogous to the OPT page replacement policy for virtual memory systems; both policies are tools for computing upper bounds on memory system performance. In this paper, perfect page placement is presented and proved NP-hard for demand-paged virtual memory systems with direct-mapped caches and infinite physical memory capacity. A class of perfect placement approximation algorithms is also presented and proved NP-hard.
求助全文
通过发布文献求助,成功后即可免费获取论文全文。 去求助
来源期刊
自引率
0.00%
发文量
0
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
确定
请完成安全验证×
copy
已复制链接
快去分享给好友吧!
我知道了
右上角分享
点击右上角分享
0
联系我们:info@booksci.cn Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。 Copyright © 2023 布克学术 All rights reserved.
京ICP备2023020795号-1
ghs 京公网安备 11010802042870号
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术官方微信