Adaptive scheduling strategies for cloud‐based resource infrastructures

Abstract

ABSTRACT This paper proposes to employ linear programming algorithms for global resource scheduling to reduce the extra cost, including power consumption as well as operation expenditures, for remote resource access in a cloud‐based resource pool. Unlike previous static work in this field, the proposed scheduler adapts the problem‐modeling granularity and resolution algorithm to the changing demands of an integral procedure comprising various stages including the initial construction and subsequent operation/extension of a cloud‐based resource infrastructure. In particular, the proposed scheduling strategies take into account resource configuration, service deployment and real‐time load, among other factors, to strike a trade‐off among the scheduling performance (i.e., cost reduced), response time, and computation cost, from a service or infrastructure operator's point of view. Copyright © 2012 John Wiley & Sons, Ltd.