A New Design for Smart and Decentralized Approach for Resource Allocation in Cloud Computing using CSP Model

Abstract

Cloud computing is the latest technology in both hardware and software. The cloud computing architecture offers a number of services including the SAAS, PAAS and IAAS. When customers specify their technical and economic needs, cloud computing helps them satisfy their request at the lowest possible cost and the best quality of resources. This paper presents a new approach for resource allocation (RA) in cloud computing. Its architecture is based on multi-agent system (MAS) and distributed constraint satisfaction problems (DCSP) where variables and constraints are distributed among multiple agents. In MAS, each agent makes his choice via a distributed negotiation to satisfy the global objective which is looking for the best solution to the users' needs. The present work highlights a new approach that is based on hybridizing (DCSP) and (MAS). This decentralized approach is designed to solve a number of problems such as resource allocation and planning cloud-computing systems.