Research And Application About Ditributed Transaction Process Protocol

Distributed transaction management is an Important function of distribute database.Its primary target is to maintain the properties of the distributed transaction,such as Atomicity,Consistency,Isolation and Durability.The two-phase commit protocol is widely applied in the practical applications for ensuring ACID nature of the transaction commitment.Two-phase commit protocol is both simple and compact,easy to use,but when its experience site failure or unsmooth network communications in the complex network environment of large distributed transaction processing may be prone to be blocking state,affect system performance and the efficiency of system resource utilization.Therefore,how to reduce the two-phase commit protocol's blocking probability has an important significance to the distributed database system.This paper introduces the related concepts of the transaction commit protocol,the goal of the distributed transaction management and recovery of Distributed Transaction.This paper described the mechanism of the common commit transaction,and analyzes the advantage and disadvantage of them.This paper focuses on analyzing the traditional communication structure of two-phase commit protocol,and propose a type of improved two phase commit protocol based on hierarchical communication structure,in order to correct the shortage of two phase commit protocol which vulnerable to be blocked state.In the improved agreement adds an auxiliary coordinator,and given the auxiliary coordinator selection algorithm.The improved protocol reduces the burden of coordinator and reduce the probability of blocking at the same time.This paper has given complete algorithm,termination protocols and the recovery of the improved two-phase commit protocol.This paper analysis the differences in fault treatment between the improved two-phase commit protocol and the traditional two-phase commit protocol.After analysis the differences and the comparison that the result shows the improved protocol is much better in ensuring the normal operation of the distributed database system at failure state.