全部 标题 作者
关键词 摘要

OALib Journal期刊
ISSN: 2333-9721
费用:99美元

查看量下载量

相关文章

更多...

一种多核系统上基于页着色的内存管理方法

DOI: 10.3969/j.issn.1005-3026.2014.03.011, PP. 351-355

Keywords: 多核,cache,实时,页着色,操作系统

Full-Text   Cite this paper   Add to My Lib

Abstract:

当今多核平台多采用共享cache架构,但运行在不同核心上的任务产生的cache冲突问题使得程序最坏执行时间的计算变得十分困难.因此提出了使用页着色技术解决多核cache上访存冲突问题的方法.此方法的优势是使已有单核上的WCET分析技术可以对多核上的程序执行时间进行判断.在Linux系统上实现了支持页着色划分方法的内存管理系统,并使用通用测试集对该方法进行了测试.实验结果表明,在Linux系统中使用该内存管理策略后,在相同多核平台上程序的执行时间变得可预测.

References

[1]  Wilhelm R, Engblom J, Ermedahl A, et al.The worstcase executiontime problemoverview of methods and survey of tools[J].ACM Transactions on Embedded Computing Systems, 2008, 7(3):1/53.
[2]  Li Y, Suhendra V, Liang Y, et al.Timing analysis of concurrent programs running on shared cache multicores[C]//RealTime Systems Symposium.Washington D C: IEEE, 2009:57/67.
[3]  Kessler R E, Hill M D.Page placement algorithms for large realindexed caches[J].ACM Transactions on Computer Systems, 1992, 10(4):338/359.
[4]  Cho S, Jin L.Managing distributed, shared L2 caches through OSlevel page allocation[C]//IEEE/ACM International Symposium on Microarchitecture.Washington D C:IEEE, 2006:455/468.
[5]  Lin J, Lu Q, Ding X, et al.Gaining insights into multicore cache partitioning:bridging the gap between simulation and real systems[C]//High Performance Computer Architecture.Salt Lake City:IEEE, 2008:367/378.
[6]  Soares L, Tam D, Stumm M.Reducing the harmful effects of lastlevel cache polluters with an OSlevel, softwareonly pollute buffer[C]//International Symposium on Microarchitecture.Washington D C:IEEE Computer Society, 2008:258/269.
[7]  Zhang X, Dwarkadas S, Shen K.Towards practical page coloringbased multicore cache management[C]//Proceedings of the 4th ACM European Conference on Computer Systems.New York: ACM, 2009:89/102.
[8]  Jeong M K, Yoon D H, Sunwoo D, et al.Balancing DRAM locality and parallelism in shared memory CMP systems[C]//High Performance Computer Architecture.New Orleans:IEEE, 2012:1/12.
[9]  Guan N, Stigge M, Yi W, et al.Cacheaware scheduling and analysis for multicores[C]//Proceedings of the seventh ACM International Conference on Embedded Software.New York:ACM, 2009:245/254.
[10]  Yotov K, Pingali K, Stodghill P.Automatic measurement of memory hierarchy parameters[J].ACM SIGMETRICS Performance Evaluation Review, 2005, 33(1):181/192.

Full-Text

Contact Us

service@oalib.com

QQ:3279437679

WhatsApp +8615387084133