SR-IOV密码设备中断传输过程建模

doi:10.11959/j.issn.2096-109x.2019007

网络与信息安全学报 ›› 2019, Vol. 5 ›› Issue (1): 56-65.doi: 10.11959/j.issn.2096-109x.2019007

• 专栏：密码攻防技术研究及应用探索 • 上一篇下一篇

SR-IOV密码设备中断传输过程建模

孙磊,李帅,郭松辉

信息工程大学，河南郑州 450001

修回日期:2018-12-26 出版日期:2019-02-01 发布日期:2019-04-17
作者简介:孙磊（1973- ），男，江苏靖江人，博士，信息工程大学教授，主要研究方向为网络空间安全、可信计算。|李帅（1994- ），男，河南驻马店人，信息工程大学硕士生，主要研究方向为云计算安全、I/O虚拟化性能优化。|郭松辉（1979- ），男，四川乐山人，博士，信息工程大学副教授，主要研究方向为云计算安全。
基金资助:
国家重点研发计划基金资助项目(2016YFB0501900)

Modeling the interrupt transmission process of SR-IOV cryptographic device

Lei SUN,Shuai LI,Songhui GUO

Information Engineering University,Zhengzhou 450001,China

Revised:2018-12-26 Online:2019-02-01 Published:2019-04-17
Supported by:
The National Key R ＆ D Program of China(2016YFB0501900)

摘要/Abstract

摘要：

SR-IOV密码设备在虚拟化环境下执行运算时会产生大量I/O中断，导致CPU在Root模式与Non-Root模式之间频繁切换，带来巨大的系统性能开销，影响SR-IOV密码设备的运算性能。针对该问题，对SR-IOV密码设备I/O中断进行建模分析，根据密码任务先到先服务的规则构建了M/M/1排队模型，分析了影响系统性能的关键因素，并通过仿真和实验测试验证了模型的有效性，最后对系统性能进行了测试与分析。结果表明，该模型能够定量分析中断频率对中断响应时间和系统队长的影响。

关键词: 中断, SR-IOV, 密码设备, 排队模型, 性能

Abstract:

The SR-IOV cryptographic device generates a large number of I/O interrupts when performing operations in a virtualized environment,causing the CPU to frequently switch between the Root mode and the Non-Root mode,which brings huge system performance overhead and affects the operational performance of SR-IOV cryptographic device.Aiming at this problem,the I/O interrupt of SR-IOV cryptographic device was analyzed,and a M/M/1 queuing model according to the cryptographic task first-come-first-served rules was constructed.What’s more,the key factors affecting system performance were analyzed,and the validity of the model via simulation and experiment was verified,and finally the system performance was tested.The results show that the model can quantitatively analyze the impact of the interrupt frequency on the interrupt response time and system queue length.

Key words: interrupt, SR-IOV, cryptographic device, queuing model, performance

中图分类号:

TP302.7

孙磊,李帅,郭松辉. SR-IOV密码设备中断传输过程建模[J]. 网络与信息安全学报, 2019, 5(1): 56-65.

Lei SUN,Shuai LI,Songhui GUO. Modeling the interrupt transmission process of SR-IOV cryptographic device[J]. Chinese Journal of Network and Information Security, 2019, 5(1): 56-65.

图/表 8

图1

图2

图3

图4

图5

图6

图7

图8

参考文献 18

[1]	张玉清, 王晓菲, 刘雪峰 ,等. 云计算环境安全综述[J]. 软件学报, 2016,27(6): 1328-1348.
	ZHANG Y Q , WANG X F , LIU X F ,et al. An overview of cloud computing environment security[J]. Journal of Software, 2016,27(6): 1328-1348.
[2]	郭松辉, 李清宝, 孙磊 ,等. 基于 ISSM 的密码服务系统虚拟化性能建模[J]. 系统仿真学报, 2017,29(8): 1692-1701.
	GUO S H , LI Q B , SUN L ,et al. Performance modeling of cryptographic service system virtualization based on ISSM[J]. Journal of System Simulation, 2017,29(8): 1692-1701.
[3]	牛欣源 . 实时任务模型中的性能解析——中断性能研究[J]. 计算机科学与技术, 2014(2): 55-64.
	NIU X Y . Real-time task model performance analysis——interrupt performance research[J]. Transactions on Computer Science and Technology, 2014(2): 55-64.
[4]	王泽武, 孙磊, 郭松辉 . 基于滚动优化的密码云实时任务阈值调度方法[J]. 计算机应用, 2017,37(10): 2780-2786.
	WANG Z W , SUN L , GUO S H . Real-time task threshold scheduling method by rolling optimization in cryptography cloud[J]. Journal of Computer Application, 2017,37(10): 2780-2786.
[5]	DONG Y , YANG X , LI J ,et al. High performance network virtualization with SR-IOV[J]. Journal of Parallel and Distributed Computing, 2012,72(11): 1471-1480.
[6]	XU X , DAVDA B . A hypervisor approach to enable live migration with pass through sr-iov network devices[J]. Acm Sigops Operating Systems Review, 2017,51(1): 15-23.
[7]	YOUNGE A J , WALTERS J P , CRAGO S P ,et al. Supporting high performance molecular dynamics in virtualized clusters using IOMMU,SR-IOV,and GPUDirect[C]// ACM Sigplan/Sigops International Conference on Virtual Execution Environments. 2015: 31-38.
[8]	李帅, 孙磊, 郭松辉 . 减少上下文切换的虚拟密码设备中断路径优化方法[J]. 计算机应用, 2018,38(7): 1946-1950.
	LI S , SUN L , GUO S H . A method to reduce interrupt path of virtual cryptographic device with context switching[J]. Journal of Computer Application, 2018,38(7): 1946-1950.
[9]	DONG Y , YANG X , LI X ,et al. High performance network virtualization with SR-IOV[C]// IEEE International Symposium on High Performance Computer Architecture. 2010: 1471-1480.
[10]	DONG Y , XU D , ZHANG Y ,et al. Optimizing network I/O virtualization with efficient interrupt coalescing and virtual receive side scaling[C]// IEEE International Conference on Cluster Computing. 2011: 26-34.
[11]	GUAN H B , DONG Y Z , TIAN K ,et al. SR-IOV based network interrupt-free virtualization with event based polling[J]. IEEE Journal on Selected Areas in Communications, 2013,31(12): 2596-2609.
[12]	LI J , XUE S , ZHANG W ,et al. When I/O interrupt becomes system bottleneck:efficiency and scalability enhancement for SR-IOV network virtualization[J]. IEEE Transactions on Cloud Computing, 2017,PP(99): 1-1.
[13]	GORDON A , AMIT N , HAR'EL N ,et al. ELI:bare-metal performance for I/O virtualization[C]// The 17th International Conference on Architectural Support for Programming Languages Operating Systems. 2012: 411-422.
[14]	TU C C , FERDMAN M , LEE C T ,et al. A comprehensive implementation and evaluation of direct interrupt delivery[C]// ACM Sigplan/Sigops International Conference on Virtual Execution Environments. 2015: 1-15.
[15]	王泽武 . 密码服务云管理与调度技术研究[D]. 郑州:战略支援部队信息工程大学, 2018.
	WANG Z W . Research on management and scheduling technique of cryptographic service cloud[D]. Zhengzhou:Strategic Support Force Information Engineering University, 2018.
[16]	米翔 . 高性能网络虚拟化下的适应性中断调和系统[D]. 上海:上海交通大学, 2013.
	MI X . Adaptive interrupt-free system in high performance I/O virtualization[D]. Shanghai:Shanghai Jiaotong University, 2013.
[17]	唐加山 . 排队论及其应用[M]. 北京: 科学出版社, 2016.
	TANG J S . Queuing theory and its application[M]. Beijing: Science PressPress, 2016.
[18]	DUTTA Y , SETHI V . Performance analysis of cryptographic acceleration in multicore environment[M]// Quality,Reliability,Security and Robustness in Heterogeneous Networks. Springer, 2013.

SR-IOV密码设备中断传输过程建模

Modeling the interrupt transmission process of SR-IOV cryptographic device

在线阅读

pdf下载

可视化

摘要/Abstract

引用本文

使用本文

图/表 8

参考文献 18

相关文章 2

Metrics

推荐阅读 0

[1]	潘启润,黄开枝,游伟. 基于隔离等级的网络切片部署方法[J]. 网络与信息安全学报, 2020, 6(2): 96-105.
[2]	刘勤让,刘崇阳,周俊,王孝龙. 基于线性脉动阵列的卷积神经网络计算优化与性能分析[J]. 网络与信息安全学报, 2018, 4(12): 16-24.