.png){kind=icon}
.png){kind=icon}
Return
Info-based approach in distributed mutual exclusion algorithms
.png){kind=icon}
PRE
AI
DOI：10.1016/j.jpdc.2012.01.005.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
In this paper, we propose a token-based fully distributed algorithm with token-asking method for Distributed Mutual Exclusion (DME) in a computer network composed of N nodes that communicate by message exchanges. The main goal is to introduce a new class of token-based DME algorithms called info-based algorithms. In some previous algorithms, the request to enter a critical section is sent to all nodes because the token-holding node is unknown, but in this info-based algorithm some nodes know the token-holding node and lead critical section entering requests to it, directly. This algorithm uses a logical structure in the form of a wraparound two-dimensional array which is imposed on the interconnecting network. Usually, a request message for entering the critical section is sent vertically down in the array, and eventually sent to the token-holding node with the assistant of an informed-node (common node between the row consisting of the token-holding node and the column consisting of the requester node). The nodes invoking the critical section can obtain the token with fewer message exchanges in comparison with many other algorithms. Typically, the number of message exchanges is 4 root N+1 under light demand which reduces to approximately 2 message exchanges under heavy demand. A correctness proof is provided. (C) 2012 Elsevier Inc. All rights reserved.
Keywords:
Mutual exclusion
Critical section
Distributed systems
Token-based algorithm
Message exchange
Concurrency
Journal
IF:
4
Papers:
3.8K
Citations:
4.8K