In this paper we investigate the problem of synchronising transactions in an object base. An object base is a collection of objects, much the way a database is a collection of data. An object, for our purposes, consists of a collection of variables (...
William Campbell McGee
A concurrency control protocol that purports to improve concurrency in database systems by using contextual information extracted from transaction executions is presented. The contextual information is basically the potential commit order of transactions. The protocol is based on conflicts between operation sequences, rather than simply on conflicts between operations. The authors claim that the protocol provides “strictly more concurrency than the protocols based on the traditional notion of conflicts.” The paper takes as its point of departure the locking protocol of Herlihy and Weihl [1] for hybrid atomic objects. After describing this protocol and reviewing existing definitions of various flavors of atomicity (static, dynamic, hybrid, and on-line hybrid), the paper presents the new protocol in three stages: an informal motivation, a formal overview, and an implementation description. The paper concludes with a summary and an appendix giving a correctness proof for the protocol. While the outline of the paper is excellent, much of the paper is difficult to read. The main thing lacking is a simple description of the proposed concurrency control procedure. The paper does provide examples (based on unbounded queue objects), but the correctness criteria for concurrent operations are not fully explained. The claim that the protocol provides more concurrency is based on the argument that it provides a theoretically larger set of valid transaction histories, but the practical significance of this is not discussed.
Become a reviewer for Computing Reviews.
Association for Computing Machinery
New York, NY, United States
Check if you have access through your login credentials or your institution to get full access on this article.
Sign in
