DISTRIBUTED CONCURRENCY-CONTROL BASED ON LIMITED WAIT-DEPTH

The performance of high-volume transaction processing systems for busi ness applications is determined by the degree of contention for hardwa re resources as well as for data. Hardware resource requirements may b e met cost-effectively with a data-partitioned or shared-nothing archi tecture. However, the two-phase locking (2PL) concurrency control meth od may restrict the performance of a shared-nothing system more severe ly than that of a centralized system due to increased lock holding tim es. Deadlock detection and resolution are an added complicating factor in shared-nothing systems. In this paper, we describe distributed Wai t-Depth Limited (WDL) concurrency control (CC), a locking-based distri buted CC method that limits the wait-depth of blocked transactions to one, thus preventing the occurrence of deadlocks. Several implementati ons of distributed WDL which vary in the number of messages and the am ount of information available for decision making are discussed. The p erformance 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 s ubstantial lock contention levels (caused by higher degrees of transac tion concurrency), distributed WDL outperforms the other methods to a significant degree.