• 中国计算机学会会刊
  • 中国科技核心期刊
  • 中文核心期刊

计算机工程与科学

• 论文 • 上一篇    下一篇

一种高性能流式并行加密算法

费雄伟1,2,李肯立2,阳王东1,2   

  1. (1.湖南城市学院信息科学与工程学院,湖南 益阳 413000;2.湖南大学信息科学与工程学院,湖南 长沙 410082)
  • 收稿日期:2016-03-14 修回日期:2016-09-07 出版日期:2017-02-25 发布日期:2017-02-25
  • 基金资助:

    国家自然科学基金重点项目(61133005,61432005);国家自然科学基金(61370095,61472124,61572175);湖南省研究生科研创新资助项目(CX2016B065);高性能计算协同创新中心优秀研究生创新资助项目(201512);湖南省科技计划(2015SK200062,2014FJ3111);湖南省教育厅科研项目(15C0254,14C0209);益阳市指导性科技计划(益科字[2016]51号10)

A highperformance parallel encryption algorithm for streams

FEI Xiongwei1,2,LI Kenli2,YANG Wangdong1,2   

  1. (1.School of Information Science and Engineering,Hunan City University,Yiyang 413000;
    2.College of Computer Science and Electronic Engineering,Hunan University,Changsha 410082,China)
  • Received:2016-03-14 Revised:2016-09-07 Online:2017-02-25 Published:2017-02-25

摘要:

随着网络用户的数量持续增多和对安全需求的增长,以AES加密的方式对用户数据流进行加密保护得到了广泛的应用。对服务器而言,大量用户形成的数据流具有流速高和突发性强的特点,而传统的串行加密却效率低下,会造成服务失效或服务质量差,因此在目前普及的CPU+GPU异构环境的基础上,通过流水线方式组织并行AES加密,以提高加密的性能,并通过滑动窗口进行突发流量控制,以提供高质量的流加密服务。实验结果显示,所提出的异构环境下的流式AES并行加密算法能满足高速率突发性的用户数据流的流式加密的需求,提高了加密的处理速度并有效地控制了流量。
 

关键词: 用户数据流, 异构环境, 并行加密, 高级加密标准, 效率, 流量控制

Abstract:

With constant increase of network users and security requirements, protecting user data streams by the advanced encryption standard  (AES) is widely used. As for servers, data streams produced by massive users have the properties of high speed and strong burst, but traditional serial encryption can cause service unavailability or low service quality because of low efficiency. Based on the  popular central processing unit ( CPU) + graphics processing unit (GPU) environment, we parallelize AES in the form of pipeline to improve encryption performance, and control burst flows using a slipping window to offer highquality stream encryption service. Experimental results show that our proposed parallel AES in heterogeneous environment can meet the requirements of user data streams with high speed and strong burst, improve the speed of dealing, and control flows effectively.

Key words: user data flow, heterogeneous environment, parallel encryption, advanced encryption standard, efficiency, flow control