Hybrid-LRU Caching Scheme for PDRAM Hybrid Memory Architecture in Cloud Computing

Hongtianchen Xie, Gangyong Jia, Guangjie Han, Jian Wan, Yongjian Ren, Jie Huang, Xindong You
{"title":"Hybrid-LRU Caching Scheme for PDRAM Hybrid Memory Architecture in Cloud Computing","authors":"Hongtianchen Xie, Gangyong Jia, Guangjie Han, Jian Wan, Yongjian Ren, Jie Huang, Xindong You","doi":"10.1109/IIKI.2016.22","DOIUrl":null,"url":null,"abstract":"This in recent years, the mixed memory (PDRAM) which is consisted of the phase-change memory (PRAM) and the conventional memory (DRAM) has drawn a lot of attention from both industry and academia. It is a promising alternative to replace the traditional memory architecture. PDRAM has good characteristics as large capacity, good stability and non-volatile, while PRAM has disadvantages of a limited life and large access latency. Traditional policies are not sufficient to be directly applied to the PDRAM memory architecture. They cannot adapt to the new features of the hybrid architecture because of their undifferentiated operation, extensive use of PRAM can cause performance degradation and shorten the life of PDRAM memory. This paper proposed Hybrid-LRU caching scheme to address these problems. Our scheme mainly uses the cache consistency address resolution mode to distinguish between different physical mediums in PDRAM, and then take different actions depending on different physical mediums. We have taken a variety of tradition policies like LRU, FIFO, RANDOM, CFLRU to make experimental comparisons. The experimental results have shown that our scheme can reduce the PRAM utilization rate of 11.8% and improve the performance by 4.6%, energy consumption of write and read operation can be reduced up to 88.2%.","PeriodicalId":371106,"journal":{"name":"2016 International Conference on Identification, Information and Knowledge in the Internet of Things (IIKI)","volume":"234 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2016-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"1","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"2016 International Conference on Identification, Information and Knowledge in the Internet of Things (IIKI)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/IIKI.2016.22","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 1

Abstract

This in recent years, the mixed memory (PDRAM) which is consisted of the phase-change memory (PRAM) and the conventional memory (DRAM) has drawn a lot of attention from both industry and academia. It is a promising alternative to replace the traditional memory architecture. PDRAM has good characteristics as large capacity, good stability and non-volatile, while PRAM has disadvantages of a limited life and large access latency. Traditional policies are not sufficient to be directly applied to the PDRAM memory architecture. They cannot adapt to the new features of the hybrid architecture because of their undifferentiated operation, extensive use of PRAM can cause performance degradation and shorten the life of PDRAM memory. This paper proposed Hybrid-LRU caching scheme to address these problems. Our scheme mainly uses the cache consistency address resolution mode to distinguish between different physical mediums in PDRAM, and then take different actions depending on different physical mediums. We have taken a variety of tradition policies like LRU, FIFO, RANDOM, CFLRU to make experimental comparisons. The experimental results have shown that our scheme can reduce the PRAM utilization rate of 11.8% and improve the performance by 4.6%, energy consumption of write and read operation can be reduced up to 88.2%.
查看原文
分享 分享
微信好友 朋友圈 QQ好友 复制链接
本刊更多论文
云计算中PDRAM混合内存架构的Hybrid- lru缓存方案
近年来,由相变存储器(PRAM)和传统存储器(DRAM)组成的混合存储器(PDRAM)受到了业界和学术界的广泛关注。它是替代传统内存架构的一种很有前途的选择。PDRAM具有容量大、稳定性好、不易失失等优点,而PRAM的缺点是寿命有限、访问时延大。传统的策略不足以直接应用于PDRAM内存架构。它们无法适应混合架构的新特性,因为它们的无差异操作,大量使用PRAM会导致性能下降并缩短PDRAM内存的寿命。为了解决这些问题,本文提出了Hybrid-LRU缓存方案。我们的方案主要采用缓存一致性地址解析模式来区分PDRAM中不同的物理介质,然后根据不同的物理介质采取不同的操作。采用LRU、FIFO、RANDOM、CFLRU等多种传统策略进行实验比较。实验结果表明,该方案可使PRAM利用率降低11.8%,性能提高4.6%,读写操作能耗降低88.2%。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
求助全文
约1分钟内获得全文 去求助
来源期刊
自引率
0.00%
发文量
0
期刊最新文献
Research on the Evaluation of Product Quality Perceived Value Based on Text Mining and Fuzzy Comprehensive Evaluation A New Pre-copy Strategy for Live Migration of Virtual Machines Hbase Based Surveillance Video Processing, Storage and Retrieval Mutual Information-Based Feature Selection and Ensemble Learning for Classification Implicit Correlation Intensity Mining Based on the Monte Carlo Method with Attenuation
×
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
现在去查看 取消
×
提示
确定
0
微信
客服QQ
Book学术公众号 扫码关注我们
反馈
×
意见反馈
请填写您的意见或建议
请填写您的手机或邮箱
已复制链接
已复制链接
快去分享给好友吧!
我知道了
×
扫码分享
扫码分享
Book学术官方微信
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术
文献互助 智能选刊 最新文献 互助须知 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 Book学术 All rights reserved.
ghs 京公网安备 11010802042870号 京ICP备2023020795号-1