基于占用率的体系结构脆弱因子在线计算方法

J4 ›› 2014, Vol. 36 ›› Issue (05): 779-785.

• 论文 • 下一篇

基于占用率的体系结构脆弱因子在线计算方法

潘送军1,2，陈传鹏1

（1.国网湖南省电力公司信息通信公司，湖南长沙 410007;2.中国科学院计算技术研究所，北京 100190）

收稿日期:2012-09-24 修回日期:2012-12-27 出版日期:2014-05-25 发布日期:2014-05-25
基金资助:
国家973计划资助项目（2011CB302503）

An occupancy-based online architecture
vulnerability factor computing method

PAN Songjun1,2,CHEN Chuanpeng1

（1.Information & Communication Company,State Grid Hunan Electric Power Company,Changsha 410007;
2.Institute of Computing Technology,Chinese Academy of Sciences,Beijing 100190,China）

Received:2012-09-24 Revised:2012-12-27 Online:2014-05-25 Published:2014-05-25

摘要/Abstract

摘要：

随着CMOS工艺的不断进步，单个芯片上集成的晶体管数目快速增长，使得由高能粒子和α粒子辐射产生的软错误逐渐成为影响微处理器可靠性的重要因素。通过计算体系结构脆弱因子，量化分析软错误对微处理器中不同结构的影响，已成为指导微处理器容错设计、提高系统可靠性的重要方法。体系结构脆弱因子在线计算方法，利用体系结构脆弱因子随应用程序运行而变化的特点，指导动态选择容错技术，实现可靠性与性能之间的平衡。在对已有方法分析的基础上，提出一种基于占用率的体系结构脆弱因子在线计算方法，并在SimAlpha模拟器上进行验证。通过运行SPEC CPU2000基准测试程序，计算发射队列、重命名缓冲和存取队列的体系结构脆弱因子。实验结果表明，该方法能有效计算微处理器中不同结构的体系结构脆弱因子，与离线计算方法相比，发射队列、重排序缓存和存取队列三个体系结构脆弱因子的平均绝对误差仅为0.10、0.01和0.039。

关键词: 微处理器, 可靠性, 软错误, 容错, 体系结构脆弱因子

Abstract:

As CMOS technology scales into the nanometer era, more transistors can be integrated in a single chip, which makes microprocessors become more vulnerable to highenergy particle induced soft errors. Quantitatively analysis of the Architectural Vulnerability Factor (AVF) of different structures is an important method to guide the microprocessor reliable design. Online AVF computing techniques utilize the features that AVF varies significantly across different workloads and individual structures, and then make a tradeoff between system reliability and performance. The paper proposes an occupancybased online AVF computing method, and evaluates it with a cycleaccurate simulator SimAlpha. Experimental results on SPEC CPU2000 integer benchmark suite show that the proposed method can compute AVF efficiently for different structures. Compared with an offline method, the averaged absolute errors of AVF for issue queue, reorder buffer, and load/store queue are 0.10, 0.01, and 0.039 respectively.

Key words: microprocessor;reliability;soft error;fault tolerance;architecture vulnerability factor

潘送军1,2，陈传鹏1. 基于占用率的体系结构脆弱因子在线计算方法[J]. J4, 2014, 36(05): 779-785.

PAN Songjun1,2,CHEN Chuanpeng1. An occupancy-based online architecture
vulnerability factor computing method [J]. J4, 2014, 36(05): 779-785.

参考文献

［1］Shivakumar P,Kistler M,Keckler S,et al.Modeling the effect of technology trends on the soft error rate of combinatorial logic［C］∥Proc of the International Symposium of Dependable Systems and Networks (DSN’02), 2002:389398.
［2］Mukherjee S S,Weaver C,Emer J,et al.A systematic methodology to compute the architectural vulnerability factors for a high performance microprocessor［C］∥Proc of the International Symposium on Microarchitecture (MICRO’03), 2003:2940.
［3］Rotenberg E.ARSMT:A microarchitectural approach to fault tolerance in microprocessors［C］∥Proc of the FaultTolerant Computing Systems (FTCS’99), 1999:8491.
［4］Austin T. DIVA:A reliable substrate for deep submicron microarchitecture［C］∥Proc of the International Symposium on Microarchitecture (MICRO’99), 1999:196207.
［5］Vijaykumar T, Pomeranz I, Cheng K. Transientfault recovery using simultaneous multithreading［C］∥Proc of the International Symposium on Computer Architecture (ISCA’02), 2002:8798.
［6］Pan Songjun, Hu Yu, Li Xiaowei. IVF:Characterizing the vulnerability of microprocessor structures to intermittent faults［J］. IEEE Transactions on Very Large Scale Integration (VLSI) Systems, 2012, 20(5):777790.
［7］Fu X, Li T, Fortes J. SimSODA:A unified framework for architectural level software reliability analysis［C］∥Proc of Workshop on Modeling, Benchmarking and Simulation, 2006:1.
［8］Saggese G P, Wang N, Kalbarczyk Z, et al. An experimental study of soft errors in microprocessors［J］. IEEE Micro, 2005, 25(6):3039.
［9］Walcott K R, Humphreys G, Gurumurthi S. Dynamic prediction of architectural vulnerability from microarchitectural state［C］∥Proc of the International Conference of Computer Architecture (ISCA’07), 2007:516527.
［10］Li X D, Adve S V, Bose P, et al. Online estimation of architectural vulnerability factor for soft errors［C］∥Proc of the International Conference of Computer Architecture (ISCA’08), 2008:341352.
［11］Fahs B, Bose S, Crum M, et al. Performance characterization of a hardware mechanism for dynamic optimization［C］∥Proc of the International Symposium on Microarchitecture (MICRO’01), 2001:1627.
［12］Desikan R, Burger D, Keckler S, et al. Simalpha:A validated, executiondriven Alpha 21264 simulator［R］. Technical Report TR0123, Austin:Department of Computer Sciences, University of Texas at Austin, 2001.
［13］Sherwood T, Perelman E, Hamerly G, et al. Automatically characterizing large scale program behavior［C］∥Proc of the International Conference on Architectural Support for Programming Languages and Operating Systems (ASPLOS’02), 2002:4557.
［14］Soundararajan N, Parashar A, Sivasubramaniam A. Mechanisms for bounding vulnerabilities of processor structures［C］∥Proc of the International Symposium on Computer Architecture (ISCA’07), 2007:506515.

[1]	熊国杰, 张津铭, 贺光辉. 一种面向Chiplet互连的高效传输协议设计与实现[J]. 计算机工程与科学, 2023, 45(08): 1339-1346.
[2]	高文才, 陈小文. 一种基于混合加固的容软错误NoC路由器[J]. 计算机工程与科学, 2023, 45(08): 1376-1382.
[3]	孙智超, 张策, 江文倩, 刘凯卫, 范苗苗, 李文毓, 温雅菲. 故障检测率对软件可靠性影响实证分析[J]. 计算机工程与科学, 2022, 44(12): 2162-2173.
[4]	龚锐, 石伟, 刘威, 张剑锋, 王蕾. 基于NAND Flash的CPU安全启动设计与实现[J]. 计算机工程与科学, 2022, 44(06): 971-978.
[5]	李腾, 程哲, 贾东立, 贾耀清. 基于角色管理的实用拜占庭容错共识算法[J]. 计算机工程与科学, 2022, 44(02): 237-243.
[6]	袁远, 李世杰, 邢建英, 蒋句平. E级高性能计算机系统中监控分系统的挑战与设计[J]. 计算机工程与科学, 2021, 43(08): 1366-1375.
[7]	隋兵才, 孙彩霞, 王永文, 郭辉. 一种双端口发射队列及其性能优化[J]. 计算机工程与科学, 2021, 43(07): 1168-1172.
[8]	张玲. MEDA芯片菊花链的自动容错设计[J]. 计算机工程与科学, 2021, 43(07): 1192-1199.
[9]	陆平静, 赖明澈, 王博超, 常俊胜. 一种计算互连融合网络体系结构[J]. 计算机工程与科学, 2020, 42(10高性能专刊): 1737-1741.
[10]	张策1，伊文敏2，白睿1，盛晟1，徐早辉1，高天翼1，王瞰宇1，苏嘉尧1. SRGM下失效数据集效用与验证分析[J]. 计算机工程与科学, 2020, 42(06): 1012-1020.
[11]	惠子青,刘晓燕,严馨. 基于排错等待延迟的广义动态集成神经网络模型[J]. 计算机工程与科学, 2020, 42(04): 641-648.
[12]	王洁，康俊杰，周宽久. 基于FPGA的故障修复演化技术研究[J]. 计算机工程与科学, 2018, 40(12): 2120-2125.
[13]	郑露露,武继刚,陈辉. 网络可存活生成树的快速恢复算法[J]. 计算机工程与科学, 2018, 40(11): 1967-1973.
[14]	王余伟1，曹东2,施书成1. 实时操作系统CPU使用率监测的软件容错研究[J]. 计算机工程与科学, 2018, 40(08): 1337-1343.
[15]	王磊，瞿佳明. 基于协同过滤和Slope One算法的Web服务可靠性预测[J]. 计算机工程与科学, 2018, 40(08): 1390-1397.

基于占用率的体系结构脆弱因子在线计算方法

An occupancy-based online architecture
vulnerability factor computing method

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 15

编辑推荐

Metrics

本文评价

基于占用率的体系结构脆弱因子在线计算方法

An occupancy-based online architecture vulnerability factor computing method

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 15

编辑推荐

Metrics

本文评价

An occupancy-based online architecture
vulnerability factor computing method