面向高通量事务处理的事务编译技术

王冬慧; 朱涛; 钱卫宁

doi:10.3969/j.issn.1000-5641.2016.05.002

面向高通量事务处理的事务编译技术

doi: 10.3969/j.issn.1000-5641.2016.05.002

1.
华东师范大学数据科学与工程研究院, 上海 200062

基金项目:

国家863计划项目(2015AA015307); 国家自然科学基金(61432006)

详细信息

通讯作者:
钱卫宁, 男, 教授, 研究方向为可扩展事务处理和海量数据分析与挖掘. E-mail: wnqian@sei.ecnu.edu.cn.

计量
- 文章访问数: 301
- HTML全文浏览量: 16
- PDF下载量: 570
- 被引次数: 0
出版历程
- 收稿日期: 2016-07-01
- 刊出日期: 2016-09-25

Compilation techniques for high throughput transaction processing

1.
Institute for Data Science and Engineering, East China Normal University, Shanghai 200062, China

摘要

摘要: 针对内存数据库中CPU利用率不高的问题, 目前的研究工作集中在利用事务编译技术提升事务的执行效率和改进事务的并发控制以提升数据库的性能. 本文主要从以下几个方面对内存数据库的事务编译技术进行了综述. 第一, 介绍了事务处理的一般流程, 分析限制系统性能的因素. 第二, 分析了当前使用的事务编译技术, 包括即时编译技术、操作依赖分析技术和事务切片技术. 第三, 结合实例分析事务编译是如何提升数据库性能的, 介绍典型的内存数据库在事务编译方面的研究工作, 如Hekaton、VoltDB等. 最后给出了研究展望.
- 事务编译 /
- 并发控制 /
- 即时编译技术 /
- 操作依赖分析技术 /
- 事务切片技术
Abstract: Because of the problem of low utilization of CPU in memory database, the present research work is focused on improving the execution efficiency and concurrency control by transaction compilation technology to improve the performance of database. This article mainly introduces the following aspects of the memory database transaction compilation technology. First, this paper introduces the general process of transaction processing and analyzes the factors that limit the performance of the system. Second, we analyze the compilation techniques, including Just-in-time Compilation, Dependence Theory and Transaction Chopping. Third, we analyze the database and show how to improve the performance combined with the introduction of typical memory database, such as VoltDB, Hekaton and so on. Finally, the research prospects are given.
- transaction compilation concurrency control just-in-time compilation dependence theory transaction chopping /

HTML全文

参考文献(1)

[1]

[1] AILAMAKI A G, DEWITT D J, HILL M D, et al. DBMSs on a modern processor: Where does time go?[C]//Proceedings of International Conference on Very Large Data Bases. UK: VLDB. 1999: 266-277.
[2]哈索, 亚历山大cdot蔡尔. 内存数据库管理 [M]. SAP, 译. 北京: 清华大学出版社,2013.
[3] BERNSTEIN P, BRODIE M, CERI S, et al. The asilomar report on database research[J]. Acm Sigmod Record, 1998, 27(4): 44-113.
[4] GARCIA-MOLINA H, ULLMAN J D, WIDOM J. Database System Implementation [M]. USA: Prentice Hall, 2010, 132-179.
[5] ALEXANDER T, DANIEL I A. The Case for Determimism in Database Systems[J]. VLDB, 2010: 70-80.
[6] WIKIPEDIA. Just-in-time manufacturing [EB/OL]. [2016-06-01]. https://en.wikipedia.org/wiki/Just-in-time{\_]\linebreak manufacturing.
[7] DIACONU C, FREEDMAN C, ISMERT E, et al. Hekaton: SQL server's memory-optimized OLTP engine [J]. SIGMOD, 2013: 1-13.
[8]DIACONU C, ISMERT E, LARSON P A, et al. Compilation in the microsoft SQL server hekaton engine [J]. IEEE, 2014: 22-32.
[9] NEUMANN T. Efficiently compiling efficient query plans for modern hardware[J]. PVLDB, 2011, 4(9): 539-550.
[10] 吴亚鑫, 孙静. 基于数据库操作的相关性模型及应用[J].计算机工程与设计, 2011, 32(1): 183-187.
[11] SHASHA D, LLIRBAT F, SIMON E, et al. Transaction chopping: Algorithms and performance studies[J]. ACM Transactions on Database Systems (TODS), 1995, 20(3): 325-363.
[12] TU S, ZHENG W T, KOHLER E, et al. Speedy transactions in multicore in-memory databases [J]. Twenty-fourth ACM Symposium on Operating Systems Principles, 2013: 18-34.
[13] STONEBRAKER M, WEISBERG A. The VoltDB main memory DBMS[J]. IEEE DEBull, 2013, 36(2): 21-27.
[14] KALLMAN R, KIMURA H, NATKINS J, et al. H-store: A high-performance, distributed main memory transaction processing system[J]. PVLDB, 2008, 1(2): 1496-1499.
[15] THOMSON A, DIAMOND T, WENG S C, et al. Calvin: Fast distributed transactions for partitioned database systems[C]// ACM SIGMOD International Conference on Management of Data. 2012: 1-12.
[16] WANG Z G, MU S, CUI Y, et al. Scaling multicore databases via constrained parallel execution [C]//\"{O]ICAN F, KOVTRIKE G, MADDEN S. SIGMOD Conference. New York: ACM, 2016: 1643-1658.
[17] PANDIS I, JOHNSON R, HARDAVELLAS N, et al.Data-oriented transaction execution[J]. PVLDB, 2010, 3(1): 928-939.
[18] In-Memory Operational Database, SQL and Scale-Out. VoltDB[EB/OL]. [2016-06-03]. http://voltdb.com.
[19] Introduction to MemSQL. MemSQL[EB/OL]. [2016-06-03]. http://www.dbms2.com/2012/06/18/introduction-to-memsql/
[20] STONEBRAKER M, MADDEN S, ABADI D J, et al.The end of an architectural era it's time for a completerewrite[J].VLDB, 2007: 1150-1160.
[21] HELLERSTEIN J M, STONEBRAKER M, HAMILTON J. Architecture of a database system[J]. Foundations and Trends Databases, 2007, 1(2): 141-259.
[22] KEMPER A, NEUMANN T. HyPer: A hybrid OLTP{\&]OLAP main memory database system based on virtual memory snapshots[J]. ICDE, 2011: 195-206.

施引文献

资源附件(0)

访问统计