###
DOI:
计算机系统应用英文版:2015,24(1):1-8
←前一篇   |   后一篇→
本文二维码信息
码上扫一扫!
多核缓存优化技术研究综述
(1.复旦大学 软件学院, 上海 201203;2.复旦大学 上海市数据科学重点实验室, 上海 201203;3.复旦大学 并行处理研究所, 上海 201203)
Survey of Cache Techniques for Multicore Architecture
(1.Software School, Fudan University, Shanghai 201203, China;2.Shanghai Key Laboratory of Data Science, Fudan University, Shanghai 201203, China;3.Parallel Processing Institute, Fudan University, Shanghai 201203, China)
摘要
图/表
参考文献
相似文献
本文已被:浏览 2021次   下载 3293
Received:April 21, 2014    Revised:May 26, 2014
中文摘要: 随着多核技术的迅速发展, 并发处理和大批量数据操作成为主流, 而为了应对更加复杂的程序行为和愈发庞大的数据量, 缓存系统的效率也正面临着严重的挑战.如何在复杂的多核环境中更高效的使用缓存, 提高缓存响应速度和数据吞吐量一直是体系结构领域的重要课题和研究热点.针对多核环境中缓存的应用场景进行分析, 从缓存的效率, 内容和共享使用三个角度进行归纳和总结, 提出缓存应用的时延问题, 容量问题, 共享问题等具有针对性的问题, 并且对针对这些具体问题和情境的缓存优化技术进行总结和综述, 同时对缓存优化的一些新技术和新的优化角度进行探讨, 最后对多核缓存优化技术的发展前景进行展望.
Abstract:With the developing of multi-core techniques, concurrent processing and big-data operation are becoming more and more popular. Facing the complexity of program behavior and the large amount of input data, the cache system needs to explore more optimization opportunities. Improving the utizlization, efficiency and throughput of cache system is always the hotspot of architectural researches. This survey analyzes the inner details of cache techniques and concludes the important issues on multi-core cache optimization, like latency problem, capacity problem and sharing problem. Furthermore, this paper introduces some typical solutions for these issues, as well as the new techniques and new optimization aspects in recent researches. Finally, a description of possible developing directions is presented at the end of this survey.
文章编号:     中图分类号:    文献标志码:
基金项目:上海市科委科技攻关项目(13DZ1108800);国家高技术研究发展计划(863)(2012AA010901);国家自然科学基金(61370081)
引用文本:
闵庆豪,张为华.多核缓存优化技术研究综述.计算机系统应用,2015,24(1):1-8
MIN Qing-Hao,ZHANG Wei-Hua.Survey of Cache Techniques for Multicore Architecture.COMPUTER SYSTEMS APPLICATIONS,2015,24(1):1-8