An inexpensively elastic resource allocation model for platform as a service cloud computing

Abstract

With the growth in cloud computing there is additional complexity introduced in cloud systems and therefore there is a need for more efficient resource allocation. Autonomic computing is a promising approach for resource allocation in cloud computing and this approach advocates for self-managing ability whereby autonomic systems can allocate resources for their own needs without intervention from humans. In the Platform-asa- Service (PaaS) model, the platform provider requests for resources such as CPU and RAM from the infrastructure provider and ensures the end client who has requested for platform resources is allocated sufficient resources to meet their requirements. In the PaaS model, platform providers suffer volatile resource demands and high provisioning costs due to resource prediction errors and penalties that arise due to SLA violations.