事务存储：具有容错特性的并发控制机制

J4 ›› 2012, Vol. 34 ›› Issue (7): 46-53.

事务存储：具有容错特性的并发控制机制

宋伟

（并行与分布处理国防科技重点实验室，湖南长沙 410073）

收稿日期:2011-06-01 修回日期:2011-09-24 出版日期:2012-07-25 发布日期:2012-07-25
基金资助:
国家自然科学基金资助项目（60921062，61003087）

Transactional Memory:A Concurrent Control Mechanism with FaultTolerant Characteristics

SONG Wei

（National Laboratory for Parallel and Distributed Processing，Changsha 410073,China）

Received:2011-06-01 Revised:2011-09-24 Online:2012-07-25 Published:2012-07-25

摘要/Abstract

摘要：

随着多核处理器的发展，开发线程级并行成为提升应用程序执行性能的必要手段，这使得事务存储作为一种具有良好支持线程级并行前景的并行编程机制受到越来越多的关注。本文首先从事务存储系统的冲突检测机制和数据版本管理机制的角度对事务存储系统进行了分类；然后总结综述了目前主要的事务存储系统的实现方式；最后从容错的角度重新审视了事务存储，我们认为事务存储本身具有良好的容错特性，可以自然地与一些主要的容错技术结合，实现高效的故障隔离、检测及恢复。

关键词: 事务存储, 容错, 多核处理器

Abstract:

With the development of the multicore processor, developing threadlevel parallelism becomes a necessary means to improve the execution performance of the application programs. It makes transactional memory, which is a promising mechanism to support threadlevel parallelism, attract more and more attention. In this paper, firstly, we classify the transactional memory systems according to the conflict detection mechanism and the dataversioning mechanism. After that, we summarize the main implementations of the transactional memory. Finally, we reexamine the transactional memory from the perspective of faulttolerance. We believe that the transactional memory has good characteristics of faulttolerance, and it can combine with some faulttolerant techniques naturally to achieve efficient fault isolation, fault detection and fault recovery.

Key words: transactional memory;fault tolerance;multicore processor

宋伟. 事务存储：具有容错特性的并发控制机制[J]. J4, 2012, 34(7): 46-53.

SONG Wei. Transactional Memory:A Concurrent Control Mechanism with FaultTolerant Characteristics[J]. J4, 2012, 34(7): 46-53.

参考文献

［1］Gray J,Reuter A.Transaction Processing:Concepts and Techniques［M］. San Francisco:Morgan Kaufmann Publishers, CA, 1992.
［2］Herlihy M, Eliot J, Moss B. Transactional Memory: Architectural Support for LockFree Data Structures［C］∥Proc of the 20th Ann. Int’l Symp Computer Architecture,1993:289300.
［3］TOP500 Supercomputing Site［EB/OL］.［20110115］.http://www.top500.org.
［4］Ananian C S,Asanovic K,Kuszmaul B C,et al.Unbounded Transactional Memory［C］∥Proc of the 11th Int’l Symp HighPerformance Computer Architecture, 2005:316327.
［5］Moore K E, Bobba J, Moravan M J, et al. LogTM: LogBased Transactional Memory［C］∥Proc of the Twelfth IEEE Symposium on HighPerformance Computer Architecture, 2006:254265.
［6］Moravan M J, Bobba J, Moore K E, et al. Supporting Nested Transactional Memory in LogTM［C］∥Proc of ASPLOSXII’06,2006:359370.
［7］Rajwar R,Herlihy M,Lai K.Virtualizing Transactional Memory［C］∥Proc of the 32nd Annual Int’l Symp on Computer Architecture, 2004:494505.
［8］Herlihy M, Luchangco V,Moir M,et al. Software Transactional Memory for DynamicSized Data Structures［C］∥Proc of the 22nd Annual Symposium on Principles of Distributed Computing, 2003:92101.
［9］Hammond L, Carlstrom B D, Wong V,et al. Transactional Coherence and Consistency: Simplifying Parallel Hardware and Software［J］.IEEE Computer Society,2004,24(6):92103.
［10］Hammond L,Wong V,Chen M,et al.Transactional Memory Coherence and Consistency［C］∥Proc of the 31st Ann Int’l Symp Computer Architecture, 2004:102113.
［11］Fraser K. Practical LockFreedom［R］. Technical Report UCAMCLTR579, Cambridge University Computer Laboratory, 2004.
［12］Fraser K,Harris T.Concurrent Programming without Locks［J］. ACM Transaction on Computer Systems,2007,25(2):5.
［13］Shavit N, Touitou D. Software Transactional Memory［C］∥Proc of the 14th Ann ACM Symp Principles of Distributed Computing, 1995:204213.
［14］Marathe V J, Scherer III W N, Scott M L. Adaptive Software Transactional Memory［C］∥Proc of the 19th Int’l Symp Distributed Computing, 2005:354368.
［15］Harris T, Cristal A, Unsal O S, et al. Transactional Memory: An Overview［J］. IEEE Micro Special Issue: Hot Tutorials, 2007,27(3):829.
［16］Larus J, Kozyrakis C. Transactional Memory. Communications of the ACM, 2008,51(7)：8088.
［17］Herlihy M P, Moss J E B. Transactional Memory: Architectural Support for LockFree Data Structures［R］. Technical Report 92/07, Digital Cambridge Research Lab, One Kendall Square, Cambridge MA 02139, 1992.
［18］Damron P, Fedorova A, Lev Y, et al. Hybrid Transactional Memory［C］∥Proc of the 12th International Conference on Architectural Support for Programming Languages and Operating Systems, 2006:336346.
［19］Kumar S, Chu M, Hughes C J, et al. Hybrid Transactional Memory［C］∥Proc of the 11th ACM SIGPLAN Symposium on Principles and Practice of Parallel Programming, 2006:209220.
［20］Saha B,AdlTabatabai A R,Jacobson Q.Architectural Support for Software Transactional Memory［C］∥Proc of the 39th International Symposium on Microarchitecture, 2006:185196.
［21］Dubrova E. FaultTolerant Design: An Introduction［M］. Kluwer Academic Publishers, 2008.
［22］宋伟, 杨学军. 基于事务回退的事务存储系统的故障恢复［J］. 软件学报, 2011, 22(9):22482262.
［23］Song Wei, Jia Jia, Peng Yuxing. Error Detection by Redundant Transaction in Transactional Memory System［C］∥Proc of 2011 Sixth IEEE International Conference on Networking, Architecture, and Storage,2011:220224.

[1]	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.
[2]	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.
[3]	罗莉, 周宏伟, 周理, 潘国腾, 周海亮, 刘彬. 一种多核处理器直连接口QoS的设计与验证[J]. 计算机工程与科学, 2021, 43(04): 620-627.
[4]	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.
[5]	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.
[6]	陈倩,刘云,高钰莹. 并行动态位向量频繁闭合序列模式挖掘算法[J]. 计算机工程与科学, 2018, 40(10): 1717-1725.
[7]	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.
[8]	章子凯，武继刚，姜文超，刘竹松. 容错处理器阵列的多逻辑列并行重构算法[J]. 计算机工程与科学, 2018, 40(01): 24-33.
[9]	张耀国1，武继刚2，高任飞1. 具有可控冗余度的虚拟机容错分配算法[J]. 计算机工程与科学, 2017, 39(09): 1617-1626.
[10]	杨胜哲，于俊清,唐九飞. 数据流程序动态调度与优化[J]. 计算机工程与科学, 2017, 39(07): 1201-1210.
[11]	魏朝磊1，闫民2，赵方1. OpenMP多核技术在颗粒流体力学方法GHM中的应用[J]. 计算机工程与科学, 2017, 39(07): 1234-1240.
[12]	姚文军,陈俊仕,苏志超,余洋,廖陈志,安虹. 基于神威太湖之光的NAMD软件的移植与优化[J]. 计算机工程与科学, 2017, 39(06): 1022-1030.
[13]	敖章衡，张应辉，郑东. 电子医疗环境下容错且可验证的数据检索方案[J]. 计算机工程与科学, 2017, 39(06): 1063-1069.
[14]	秦宁宁1,2，吴德恩1，余颖华1. 基于骨干多边形的传感器网络分区双连通恢复算法[J]. 计算机工程与科学, 2017, 39(04): 673-677.
[15]	杜琦，姜浩，李宽，彭林，杨灿群. 面向ARMv8 64位多核处理器QTRSM的实现[J]. 计算机工程与科学, 2017, 39(03): 451-457.