The existing mutual exclusion algorithms in distributed systems can be categorized as token-based or non-token-based. Some of these algorithms focus on reducing message traffic, others on minimizing time delay. However, there is no single algorithm that can minimize message traffic and time delay simultaneously. Most algorithms reduce message traffic at the cost of increasing time delay. In this paper, the author introduces a hybrid algorithm that can reduce message traffic and time delay at the same time by combining two mutual exclusion algorithms such that one algorithm minimizes message traffic while the other minimizes time delay.

The paper has a clear and classical layout. The introduction summarizes the contents of the paper and states its purpose. A full presentation of the topic follows. The author demonstrates her innovation by designing a hybrid mutual exclusion algorithm using Singhal’s dynamic information structure algorithm to minimize time delay, and Maekawa’s algorithm to minimize message traffic. The author also indicates that the most difficult part of the design of a hybrid algorithm is controlling the interaction between the local and global algorithms. She proposes an algorithm that has an efficient way to control the interaction. This hybrid mutual exclusion algorithm uses the release-local-sites-first mode, the requesting group semantics, and four substeps (local competition followed by global competition). A simulation study indicates that the proposed algorithm can reduce message traffic by 52 percent and time delay by 29 percent. The author concludes that the reduction is a result of the successive executions of the critical section in the same group, the use of Maekawa’s algorithm among groups to reduce message traffic, and the use of Singhal’s dynamic information structure algorithm within a group to reduce time delay.

With some knowledge of mutual exclusion, this paper is easy to read and understand. It contains some interesting results that contribute to this field.