Due to the growing application of peer to peer computing, distributed applications are continuouslyspreading over an extensive number of nodes. To cope with this large number of participants, varioushierarchical cluster based solutions have been proposed. Cluster or group based solutions are scalable fora large number of participants. As far as group mutual exclusion solutions are concerned, some of themhave good complexity but do not take into account the growing number of participants. Others take intoaccount the previous aspect but do not have good complexity. In this paper we present two group mutualexclusion algorithms namely; TBGMEACα and TBGMEACβ to deal both with the growing number ofnodes and the matter of having a good complexity. The proposed logical structure is a cluster tree. The firstsolution uses the partial flooding method, which is the partial propagation of informations available at rootlevel. In the second solution, informations available at root level are propagated only towards theprocesses which issued a request for a session. Both algorithms have complexities of O(p) and O(log p)respectively (p is the number of clusters in the system).
展开▼
