Transmission Power Consumption Laxity-Based (TPCL) Algorithm for Communication Type Applications

Abstract

In order to realize energy-aware information systems, it is critical to discuss how to reduce the total electric power consumption of information systems. In this paper, we consider communication type applications where a server transmits a large volume of data to clients. A client first selects a server in a cluster of servers and issues a file transmission request to the server. In order to realize the server selection algorithm, it is critical to define a power consumption model of a server to perform processes. In our previous studies, the transmission power consumption (TPC) model and the extended TPC (ETPC) model of a server to transmit files to clients are proposed. In the TPC model, the electric power consumption of cooling devices like fans is assumed to be constant. Thus, the total power consumption of a server depends on not only computation and communication devices but also cooling devices. In the ETPC model, the power consumption of cooling devices is considered. Each server can be modeled as TPC or ETPC depending on cooling devices. In this paper, we newly propose the transmission power consumption laxity-based (TPCLB)algorithm for a cluster of servers which are based on the TPC and ETPC models so that the total power consumption in a cluster of servers can be reduced. We evaluate the TPCLB algorithm in terms of the total power consumption and elapse time compared with the round-robin (RR) algorithm.