ABSTRACT:
There is a growing awareness among entrepreneurs of the need to be IT compliant. It is imperative to ensure that agreed Service Level Agreements (SLAs) are respected as customer satisfaction remains a vital focus of Service Providers (SPs). Appraisal of the minimum quantity of resources that a service provider has to provide to guarantee stipulated SLAs deserves a lot of attention and remains a viable research area. Hence, this study was set to implement a monitoring system for managing cloud service disposition based on acceptable SLAs.
A purposive design for determining in advance the needed Virtual Machines (VMs) to satisfy specified Quality of Service (QOS) in the SLAs was setup. This design proposed techniques for optimal provisioning of Cloud resources with the aim of maximizing profit by handling the dynamism associated with SLAs and heterogeneous resources. The proposed technique were evaluated using the CloudSim simulator with workloads from CloudMinder (a service product from Computer Associates (CA) Technologies). Data centres with physical machines which configuration resembled public Cloud such as Amazon EC2 large image, were simulated. The VMs of different types were mapped to physical machines and the general scheduling policy was time shared. The negotiation framework performs an adaptive and intelligent bilateral bargaining of SLAs between SaaS brokers and SaaS providers necessary for effective decisions for negotiation.
A prototype system called Service Level Agreement Management System (SLAMS) was implemented using SharePoint platform and Java programming language to validate and demonstrate the usefulness and feasibility of the proposed technique.
In conclusion, the negotiation framework provided more flexibility in terms of services needed to cater for variations associated with an individual customer. The use of the proposed framework for resource management by enterprise software as service providers was therefore recommended.
Keywords: Cloud Computing, Resources, Service Level Agreement, Software Service, Quality
