Service Composition Using Efficient Multi-agents in Cloud Computing Environment

Abstract

In today’s Internet world, end users need everything as a service (EaaS) being available to them throughout the world. Keeping users in the centre, developers are emplaning cloud computing environment which can satisfy needs of users virtually. Cloud computing is a collection of Web-accessible resources (i.e. Web services) that should be dynamically composed between service providers and brokers and virtualized based on consumer’s needs on an on-demand basis. Mapping of the users’ requirements should be done in an automated manner. But, distributed and constantly changing cloud computing environments pose new challenges to automated service composition such as: (i) dealing with incomplete information regarding cloud resources (e.g. location and providers), and (ii) dynamics of service providers, which set service fees on a supply-and-demand basis. To address these issues, we have proposed a multi-agent-based approach to compose services in multi-cloud environments for different types of cloud services: one-time virtualized services, persistent virtualized services, vertical services, and horizontal services. Cloud participants and resources are implemented and instantiated by agents. Previously the researchers have proposed self-organizing agents those make use of Service Capability Table and the semi-recursive contract net protocol (SR-CNP) to evolve and adapt cloud service compositions. To the existing work, we have planned to modify some of agents’ behaviours, as a result we can reduce number of message passing by half in order to increase overall performance. Also we are planning a 2-layered (3 levels of multi-agents) self-organizing MAS that will establish a cloud service composition.