Lamport's distributed mutual exclusion algorithm
- Every process maintains a queue of pending requests for entering critical section in order. The queues are ordered by virtual time stamps derived from Lamport timestamps.
- Pushing its request in its own queue (ordered by time stamps)
- Sending a request to every node.
- Waiting for replies from all other nodes.
- If own request is at the head of its queue and all replies have been received, enter critical section.
- Upon exiting the critical section, remove its request from the queue and send a release message to every process.
- After receiving a request, pushing the request in its own request queue (ordered by time stamps) and reply with a time stamp.
- After receiving release message, remove the corresponding request from its own request queue.
This algorithm creates 3(N − 1) messages per request, or (N − 1) messages and 2 broadcasts. 3(N − 1) messages per request includes:
- (N − 1) total number of requests
- (N − 1) total number of replies
- (N − 1) total number of releases
This algorithm has several disadvantages. They are:
- It is very unreliable as failure of any one of the processes will halt progress.
- It has a high message complexity of 3(N - 1) messages per entry/exit into the critical section.
- Kshemkalyani, A., & Singhal, M. Chapter 9: Distributed Mutual Exclusion Algorithms. Distributed Computing: Principles, Algorithms, and Systems (Page 10 of 93). Cambridge University Press.