Adaptive Admission Control Policy in Distributed Real Time Database System

Abstract

Today's real time computers activities and applications required data services in the distributed platform where real time applications have been needed for databases, in which data and operations have special characteristics in terms of time constraints. It stores large amount of data with time-sensitive manner and process it within fix given time frame. The requirement for real time computers activities and its applications are increasing in distributed real time database system (DRTDBS) including advanced traffic control, global environment control, nationwide power grid control etc. The transaction work load in DRTDBS has not balanced and the transaction processing patterns have time varying and distorted. Due to the unbalanced computation workload on DRTDBS and unpredictable communication delay can lead to miss the large number of transaction deadline. Hence, an adaptive admission control architecture and protocol have required for DRTDBS. The Adaptive admission Control Policy has fulfil to accessing and maintaining real time computers activities and application of data for satisfy timing constraints of transaction supported applications. This paper presents an adaptive admission control architecture and an efficient set of protocol. The local controller controls the local transaction request and global load balancer works global transaction request on distributed environment. The functionality of Adaptive Admission Control Policy have increased the performance because minimizing the miss ratio and increasing throughput. The experiment results show that an adaptive admission control policy effectively balances the workload in distributed real time database system under tight transaction workload as compared to existing protocols.