A Demand Based Load Balanced Service Replication Model

Abstract

Cloud computing allows service users and providers to access the applications, logical resources and files on any computer with ease. A cloud service has three distinct characteristics that differentiate it from traditional hosting. It is sold on demand, typically by the minute or the hour; it is elastic. It is a way to increase capacity or add capabilities on the fly without investing in new infrastructure, training new personnel, or licensing new software. It not only promises reliable services delivered through next-generation data centers that are built on compute and storage virtualization technologies but also addresses the key issues such as scalability, reliability, fault tolerance and file load balancing. The one way to achieve this is through service replication across different machines coupled with load balancing. Though replication potentially improves fault tolerance, it leads to the problem of ensuring consistency of replicas when certain service is updated or modified. However, fewer replicas also decrease concurrency and the level of service availability. A balanced synchronization between replication mechanism and consistency not only ensures highly reliable and fault tolerant system but also improves system performance significantly. This paper presents a load balancing based service replication model that creates a replica on other servers on the basis of number of service accesses. The simulation results indicate that the proposed model reduces the number of messages exchanged for service replication by 25-55% thus improving the overall system performance significantly. Also in case of CPU load based file replication, it is observed that file access time reduces by 5.56%-7.65%.