Distributed Concurrency Control Based on Limited Wait-Depth

The performance of high-volume transaction processing systems for business applications is determined by the degree of contention for hardware resources as well as for data. Hardware resource requirements may be met cost-effectively with a data-partitioned or shared-nothing architecture. However, the two-phase locking (2PL) concurrency control method may restrict the performance of a shared-nothing system more severely than that of a centralized system due to increased lock holding times. Deadlock detection and resolution are an added complicating factor in shared-nothing systems. The authors describe distributed Wait-Depth Limited (WDL) concurrency control (CC), a locking-based distributed CC method that limits the wait-depth of blocked transactions to one, thus preventing the occurrence of deadlocks. Several implementations of distributed WDL which vary in the number of messages and the amount of information available for decision making are discussed. The performance of a generic implementation of distributed WDL is compared with distributed 2PL (with general-waiting policy) and the Wound-Wait CC method through a detailed simulation. It is shown that distributed WDL behaves similarly to 2PL for low lock contention levels, but for substantial lock contention levels (caused by higher degrees of transaction concurrency), distributed WDL outperforms the other methods to a significant degree. >

[1]  Daniel J. Rosenkrantz,et al.  System level concurrency control for distributed database systems , 1978, TODS.

[2]  J CareyMichael,et al.  Concurrency control performance modeling: alternatives and implications , 1987 .

[3]  Y. C. Tay,et al.  Locking performance in centralized databases , 1985, TODS.

[4]  Erhard Rahm,et al.  A new distributed optimistic concurrency control method and a comparison of its performance with two-phase locking , 1990, Proceedings.,10th International Conference on Distributed Computing Systems.

[5]  Alexander Thomasian.  Performance analysis of locking policies with limited wait depth , 1992, SIGMETRICS '92/PERFORMANCE '92.

[6]  Kenneth C. Sevcik,et al.  Comparison of Concurrency Control Methods Using Analytic Models , 1983, IFIP Congress.

[7]  Miron Livny,et al.  Concurrency control performance modeling: alternatives and implications , 1987, TODS.

[8]  Alexander Thomasian,et al.  Wait depth limited concurrency control , 1991, [1991] Proceedings. Seventh International Conference on Data Engineering.

[9]  Michael Stonebraker,et al.  The Case for Shared Nothing , 1985, HPTS.

[10]  Tom W. Keller,et al.  Locking Performance in a Shared-Nothing Parallel Database Machine , 1989, IEEE Trans. Knowl. Data Eng..

[11]  Theo Härder,et al.  Observations on optimistic concurrency control schemes , 1984, Inf. Syst..

[12]  Alexander Thomasian,et al.  Concurrency control for high contention environments , 1992, TODS.

[13]  Alexander Thomasian,et al.  Performance Analysis of Two-Phase Locking , 1991, IEEE Trans. Software Eng..

[14]  Alexander Thomasian,et al.  Determining the Number of Remote Sites Accessed in Distributed Transaction Processing , 1993, IEEE Trans. Parallel Distributed Syst..

[15]  Philip Heidelberger,et al.  A Performance Comparison of Multimicro and Mainframe Database Architectures , 1988, IEEE Trans. Software Eng..

[16]  Alexander Thomasian,et al.  Access invariance and its use in high contention environments , 1990, [1990] Proceedings. Sixth International Conference on Data Engineering.

[17]  Jim Gray,et al.  The cost of messages , 1988, PODC '88.

[18]  Miron Livny,et al.  Distributed Concurrency Control Performance: A Study of Algorithms, Distribution, and Replication , 1988, VLDB.

[19]  John T. Robinson,et al.  Limitations of concurrency in transaction processing , 1985, TODS.