An online fuzzy Decision Support System for Resource Management in cloud environments

Abstract

Cloud computing is a large-scale distributed computing paradigm driven by economies of scale, in which a pool of abstracted, virtualized, dynamically-scalable, managed computing power, storage, platforms, and services are delivered on demand to external customers over the Internet. Although a significant amount of studies have been developed to optimize resource management and task scheduling in cloud computing, none of them considered the impact of task scheduling patterns on resource management and vice versa. To overcome this drawback, and considering the lack of resources in cloud environments and growing customer demands for cloud services, this paper proposes an Online Resource Management Decision Support System (ORMDSS) that addresses both tasks scheduling and resource management optimization in a unique system. In addition, ORMDSS contains a fuzzy prediction method for predicting VM workload patterns and VM migration time by applying neural networks and fuzzy expert systems. This ORMDSS helps cloud providers to automatically allocate scare resources to the applications and services in an optimal way. It is expected that the ORMDSS not only increases cloud utilization and QoS, but also decreases cost and response time.