Optimization of resource usage in virtualized environments

Abstract: Today’s markets are heavily concentrated on Cloud Computing, where trends project an increasing amount of services living therein. Because these data centers, named Clouds, provide ability to dynamically scale the resources assigned to a service. It is a technology which allows products to remain competitive in a constantly changing market. Consumers profit from this solution, as it is much easier to maintain the quality of a given service (QoS) by allocating more resources within a Cloud, on the contrary it lies within the Cloud provider’s interest to fit as many services possible within their Cloud. This poses the question of how resources integral to facilitation of a service are meant to be allocated to enable efficient utilization of resources within the Cloud. Services consisting of several Virtual machines are placed on as few physical machines as possible using a 3-dimensional Multiple Subset-sum problem model approach. The virtual machine placement case considered is Infrastructure as a Service (IaaS), and its implementation is facilitated by Java Constraint Programming Library (JaCoP) developed by Prof. Krzysztof Kuchcinski and PhD Radosław Szymanek. Results from this work deem constraint programming a suitable approach for enhancing an existing virtual machine deployment process in an IaaS Cloud.