期刊文献+

高性能LTE终端通信协议栈缓存机制 被引量:1

CACHE MECHANISM OF HIGH PERFORMANCE LTE TERMINAL COMMUNICATION PROTOCOL STACK
下载PDF
导出
摘要 针对LTE终端通信协议栈的处理过程中Cache缓存方式效率差,提出一种可以提高处理速度降低延时的数据缓存方式。通过利用便签式存储器(Scratch-Pad Memory,SPM)和直接内存存取方式(Direct Memory Access,DMA)的性能优势,将协议栈处理的过程中需要与处理器频繁交互的数据缓存到SPM中,实现在SPM中完成数据的包头压缩/解压、加密/解密和重组等协议功能处理,并通过DMA方式实现SPM与主存之间的数据搬运,减少处理器对大块连续数据的搬运,提高处理效率。该方式避免了CPU在通过Cache缓存方式进行数据交互容易发生缓存不命中的问题,从而提高了协议栈的数据处理效率。经过理论分析及实验对比,结果表明,在LTE协议栈处理过程中,采用SPM与DMA结合的数据缓存机制,相比Cache缓存方式可以使整体性能至少提升12.65%。 Aiming at the inefficiency of the Cache buffering,in the processing of the LTE terminal communication protocol stack,this paper proposes a data buffering method which can improve the processing speed and reduce the delay.By using the performance advantages of SPM and DMA,the data which needed to exchange with the processor frequently in the process of protocol stack processing was cached in SPM.The protocol function processing such as packet head compression/decompression,encryption/decryption and reorganization were realized in SPM.The data transfer between SPM and main memory was realized by DMA,which could reduce the transfer of large continuous data and improve the processing efficiency.This method avoids the problem that CPU is prone to cache miss in data exchange through cache,and improves the data processing efficiency of protocol stack.After theoretical analysis and experimental comparison,the results show that in the process of LTE protocol stack,the data cache mechanism combined with SPM and DMA can improve the overall performance by at least 12.65%compared with the Cache mode.
作者 于欣峰 梁利平 Yu Xinfeng;Liang Liping(Institute of Microelectronics of The Chinese Academy of Science,Beijing 100029,China;School of Electrical,Electronics and Communication Engineering,University of Chinese Academy of Sciences,Beijing 100029,China)
出处 《计算机应用与软件》 北大核心 2020年第3期96-100,137,共6页 Computer Applications and Software
基金 国家自然科学基金项目(61471354)。
关键词 LTE 协议栈 缓存空间管理 DMA 便签式缓存 LTE Protocol stack Memory management DMA SPM
  • 相关文献

参考文献7

二级参考文献38

  • 1许祥,罗宇.一种SAN环境下集群文件系统的元数据缓存研究[J].计算机研究与发展,2012,49(S1):240-244. 被引量:3
  • 2周功业,吴伟杰,陈进才.一种基于对象存储系统的元数据缓存实现方法[J].计算机科学,2007,34(10):146-148. 被引量:4
  • 3Chun Liu, Anand Sivasubramaniam, Mahmut Kandemir. Organizing the Last Line of Defense before Hitting the Memory Wall for CMPs [C]//HPCA - 10, 2004.
  • 4Yuanyuan Zhou, James Philbin, Kai Li. The Muhi-Queue Replace- ment Algorithm for Second Level Buffer Caches [ C ]//2002 USENIX ATC, June 25 -30, 2001:91 - 104.
  • 5Stone H S, Turek J,Wolf J L. Optimal Parlitioning of Cache Memory[J]. IEEE Trmls. Comput,1992,41(9) :1054 - 1068.
  • 6Suh G E, Rudolph L, Devadas S. Dynamic Partitioning of Shared Cache Memory[ J ]. J. Supercomput,2004,28( 1 ) :7 - 26.
  • 7Ranjith Subramanian , Yannis Smaragdakis , Gabriel H Loh. Adaptive Caches : Effective Shaping of Cache Behavior to Workloads [ C ]//MI- CRO - 39, 2006.
  • 8Dybdahl H,Stenstrom P. An Adaptive Shared/Private NUCA Cache Partitioning Scheme for Chip Muhiprocessors[ C ]//HPCA-13, 2007.
  • 9George Alm6si, C~lin Caeeaval,et al. Dissecting Cyclops: a detailed a- nalysis of a muhithreaded architecture[ J]. ACM SIGARCH Computer Architecture News,2003,31 ( 1 ).
  • 10Qureshi M K, Jaleel A, Patt Y N, et al. Adaptive insertion policies for high performance caching[ C ]//ISCA-34, 2007.

共引文献26

同被引文献13

引证文献1

二级引证文献2

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部