Min_c: heterogeneous concentration policy for power aware scheduling

In this paper, we address power-aware online scheduling of jobs with resource contention. We propose an optimization model and present a new approach to resource allocation based on job concentration. We take into account different types of applications and heterogeneity of workloads that could include CPU-intensive, disk-intensive, I/O-intensive, memory-intensive, network-intensive and other applications. When jobs of one type are allocated to the same resource, they may create a bottleneck and resource contention either in CPU, memory, disk or network. It may result in system performance degradation and increasing energy consumption. The main objective is to minimize the total energy consumption of running heterogeneous workloads. We focus on energy characteristics of applications assuming that applications of different types contribute differently to the total power consumptions due to use different hardware. We propose a nonlinear hybrid model of energy consumption. Our model takes into account power consumption of individual jobs and their combinations. We propose heterogeneous job consolidation algorithms and validate them by conducting a performance evaluation study using the CloudSim toolkit under different scenarios and real data. We analyze several scheduling algorithms depending on the type and amount of information they require. We show that information about resources utilization without knowledge of jobs types does not help much to improve the total energy consumption. In the other hand, being aware of types of applications, intelligent allocation strategies can further improve energy consumption compared with traditional approaches.