Multi-objective genetic algorithm for multi-cloud brokering

Abstract

Cloud computing is a model for enabling ubiquitous, convenient, on-demand network access to a shared pool of configurable computing resources offered by commercial providers according to specific service level agreements. Research effort has been spent to address the lack of Cloud interoperability that is a barrier to cloud-computing adoption because of the vendor lock-in problem. In fact the ability to easily move workloads and data from one cloud provider to another or between private and public clouds can improve performance, availability and reduce costs. In this paper we explore the potential use of multiobjective genetic algorithms in the field of a brokering service, whose aim is to acquire resources from multiple providers on the basis of SLA evaluation rules finding the most suitable composition of Cloud offers that satisfy users' requirements. © 2014 Springer-Verlag Berlin Heidelberg.