Carleton University
Technical Report TR-97-04
January 1997
TR-97-04: A Performance Study of Locking Granularity in Shared-Nothing Parallel Database Systems
Sivarama P. Dandamudi, S. L. Au, C. Y. Chow
Abstract
Locking granularity refers to the size of a lockable data unit, called a “granule”, in a database system. Fine granularity improves system performance by increasing concurrency level but it also increases lock management overhead. Coarse granularity, on the other hand, sacrifices system performance but low-ers the lock management cost. This paper explores the impact of gran-ule size on performance of shared-nothing parallel database sys-tems. We also report the interaction between transaction and sub-transaction scheduling policies and the locking granularity.