DRAM中访问延迟的优化

Azhar Qazi, Z. Ullah, K. Rehman, Muhammad Hilal Khan, Muhammad Bilal
{"title":"DRAM中访问延迟的优化","authors":"Azhar Qazi, Z. Ullah, K. Rehman, Muhammad Hilal Khan, Muhammad Bilal","doi":"10.1109/ICECUBE.2016.7495216","DOIUrl":null,"url":null,"abstract":"Modern digital systems, which involve high data computations, suffer from high memory access latency; thus, latency becomes a core issue in the performance enhancement of these advance digital machines. Different factors are behind the high latency of advance digital systems. Approaches like array binding and allocation, code rewriting, and others are adopted to reduce the overall latency of these systems. In this paper, we explore new dimensions to achieve maximum latency optimization in applications that involve extensive memory access. The proposed algorithm of idle/slack time management utilizes empty slots in memory access of different memory modules by appropriately activating upcoming commands in advance. The optimization of latency is further increased by incorporating the multi-way conflict resolution algorithm in second stage and followed by the use of advance dynamic buffers in third stage. Our successive three stages approach of adopting slack time management, multi-way partitioning using min-cut algorithm, and the use of advance dynamic buffers yields better results. Comparison of the experimental results with different benchmarks shows that our proposed technique optimizes the existing page-mode technique by 9%. Hence, adopting this proposed optimization strategy significantly reduces overall latency of modern digital systems.","PeriodicalId":377662,"journal":{"name":"2016 International Conference on Computing, Electronic and Electrical Engineering (ICE Cube)","volume":"98 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2016-04-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"Optimization of access latency in DRAM\",\"authors\":\"Azhar Qazi, Z. Ullah, K. Rehman, Muhammad Hilal Khan, Muhammad Bilal\",\"doi\":\"10.1109/ICECUBE.2016.7495216\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"Modern digital systems, which involve high data computations, suffer from high memory access latency; thus, latency becomes a core issue in the performance enhancement of these advance digital machines. Different factors are behind the high latency of advance digital systems. Approaches like array binding and allocation, code rewriting, and others are adopted to reduce the overall latency of these systems. In this paper, we explore new dimensions to achieve maximum latency optimization in applications that involve extensive memory access. The proposed algorithm of idle/slack time management utilizes empty slots in memory access of different memory modules by appropriately activating upcoming commands in advance. The optimization of latency is further increased by incorporating the multi-way conflict resolution algorithm in second stage and followed by the use of advance dynamic buffers in third stage. Our successive three stages approach of adopting slack time management, multi-way partitioning using min-cut algorithm, and the use of advance dynamic buffers yields better results. Comparison of the experimental results with different benchmarks shows that our proposed technique optimizes the existing page-mode technique by 9%. Hence, adopting this proposed optimization strategy significantly reduces overall latency of modern digital systems.\",\"PeriodicalId\":377662,\"journal\":{\"name\":\"2016 International Conference on Computing, Electronic and Electrical Engineering (ICE Cube)\",\"volume\":\"98 1\",\"pages\":\"0\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2016-04-11\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"2016 International Conference on Computing, Electronic and Electrical Engineering (ICE Cube)\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.1109/ICECUBE.2016.7495216\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"2016 International Conference on Computing, Electronic and Electrical Engineering (ICE Cube)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/ICECUBE.2016.7495216","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 0

摘要

现代数字系统,涉及高数据计算,遭受高内存访问延迟;因此,延迟成为这些先进数字机器性能增强的核心问题。先进数字系统的高延迟背后有不同的因素。采用数组绑定和分配、代码重写等方法来减少这些系统的总体延迟。在本文中,我们探索了在涉及大量内存访问的应用程序中实现最大延迟优化的新维度。提出的空闲/空闲时间管理算法通过提前适当激活即将到来的命令,利用不同内存模块的内存访问空槽。第二阶段采用多路冲突解决算法,第三阶段采用先进动态缓冲区,进一步提高了延迟的优化。我们的连续三阶段方法采用空闲时间管理,使用最小切割算法的多路分区,并使用先进的动态缓冲区产生了更好的结果。与不同基准测试的实验结果比较表明,我们提出的技术比现有的页面模式技术优化了9%。因此,采用该优化策略可显著降低现代数字系统的整体延迟。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
Optimization of access latency in DRAM
Modern digital systems, which involve high data computations, suffer from high memory access latency; thus, latency becomes a core issue in the performance enhancement of these advance digital machines. Different factors are behind the high latency of advance digital systems. Approaches like array binding and allocation, code rewriting, and others are adopted to reduce the overall latency of these systems. In this paper, we explore new dimensions to achieve maximum latency optimization in applications that involve extensive memory access. The proposed algorithm of idle/slack time management utilizes empty slots in memory access of different memory modules by appropriately activating upcoming commands in advance. The optimization of latency is further increased by incorporating the multi-way conflict resolution algorithm in second stage and followed by the use of advance dynamic buffers in third stage. Our successive three stages approach of adopting slack time management, multi-way partitioning using min-cut algorithm, and the use of advance dynamic buffers yields better results. Comparison of the experimental results with different benchmarks shows that our proposed technique optimizes the existing page-mode technique by 9%. Hence, adopting this proposed optimization strategy significantly reduces overall latency of modern digital systems.
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:604180095
Book学术官方微信