An extended improved redundant power consumption laxity-based (EIRPCLB) algorithm for energy efficient server cluster systems

摘要

In cloud computing systems, server cluster systems are used to provide flexible, scalable, and fault-tolerant application services. One way to provide a fault-tolerant application service is that multiple replicas of each application process are performed on multiple servers in a server cluster. However, a large amount of electric energy is consumed in a server cluster. Hence, it is critical to discuss how to make information systems not only fault-tolerant but also energy-efficient. In our previous studies, the extended improved redundant power consumption laxity-based (EIRPCLB) algorithm is proposed to reduce the total energy consumption of a server cluster to redundantly perform application processes. Once a replica successfully terminates on one server, replicas being or to be performed on other servers are meaningless. In the EIRPCLB algorithm, the total energy consumption of a server cluster can be reduced by forcing meaningless replicas to terminate and differentiating the starting time of each replica. In this paper, we evaluate the EIRPCLB algorithm in terms of total energy consumption and the average response time in homogeneous and heterogeneous clusters. We make clear how the total energy consumption of a server cluster and response time of each process change according to the change of inter-arrival time of request processes, inter-request time of replicas, redundancy of each process, and delay time between servers.