Effective trust frame work and selection model for collaborated cloud computing service

Abstract

Cloud Computing is a service model in which end users easily newlineaccess the shared resource pool through Internet and offers effective services newlineto all consumers in an efficient way. A single cloud is insufficient to serve newlinethe requirements of numerous customers especially accessing during peak newlinetime. So, collaborated cloud is needed. Collaborated cloud is framed by newlineinterconnection of multiple cloud providers. Selecting an optimal and newlinetrustworthy provider is a challenging task for the user. Trust is a mental factor newlinewhich changes over time and it cannot be completely calculated using trust newlinemeasurement attributes. So, a combined architecture which provides newlinetrustworthy and fault tolerant services is needed in which services are newlineprovided to cloud users through cloud broker from multiple cloud service newlineproviders. Cloud broker is an agent who maps the requirement of the user newlinewith the cloud provider. In federated cloud service environment, the provided newlineservice should satisfy the QoS mentioned in SLA.One of the critical factors in success of cloud provider is the delivery of services, which enabled with promised QoS. SLA is a Service Level Agreement, which consists of SLO (Service Level Objectives) a set of newlineobjectives based on users need on the federated system like certain basic QoS newlinemetrics and some high level QoS requirements. Both cloud users and newlineproviders should be satisfied with SLA which acts as a common newlineunderstanding deed between the customers expectation and provider s newlineservice. The failure of providing the service with promised QoS is known as newlineSLA violation. The violation in these policies will lead to lack of trust newlinebetween each other like loss of goodwill, money and reputation. Hence, SLA newlineneeds to be handled efficiently for best results. The legal agreement consists newlineof set of obligations SLO, set of deliverability and SLA violation penalties. newlineTo manage SLA violations, a new architecture called Effective newlineTrust framework and Selection Model is proposed which provides efficient newlinefault tolerant techniques such as FDSSS, virtual clustering technique using newlinecustomized Bee s Algorithm and Markov chain model based resource newlineprediction technique (FCMRP). This framework can guarantee the availability newlineand reliability of service in spite of failures in the system. In the proposed newlinearchitecture, Fault tolerant Designer for SLA Satisfied Service (FDSSS) newlinecomponent is introduce newline newline