Energy-Aware Cloud Workflow Applications Scheduling With Geo-Distributed Data

Electricity prices differ during different time periods and change from place to place. Cloud workflow applications often require geo-distributed data which is transmitted among heterogeneous servers in intra- and inter- data centers. Such varying electricity prices and data transmission time bring great challenges when optimizing the energy cost for scheduling tasks in workflow applications to heterogeneous servers in cloud data centers. In this paper, we minimize the total electricity cost in a deadline constrained energy-aware workflow scheduling problem with data being geographically distributed across data centers. A scheduling framework is proposed. Strategies are developed to sequence workflow applications, divide deadlines and sort tasks. An adaptive local search method is presented to improve solutions during the search process which dynamically balances intensification using neighborhood structures of increasing size. Components and parameter values are statistically calibrated over a comprehensive set of random instances. The proposed algorithm is compared to modified classical algorithms for similar problems. Experimental results demonstrate the effectiveness of the proposal for the considered problem.