Abstract
As the multitude and complexity of cloud market increases the evaluation and selection of cloud services becomes a burdensome task for the users. With the increased rise of available services from various Cloud Service Providers (CSP), the role of cloud brokers becomes more and more important. In this thesis, the challenge of optimally allocating multiple cloud system resources to multiple mobile user’s requests with different requirements is investigated and an optimal Cloud Broker model is proposed. The cloud brokering mechanism is formulated as a Semi-Markov Decision Process (SMDP) model under the average system cost criteria, taking into consideration the cost of the occupying computing resources, the communication costs, the request traffic, and some security risk degrees and resource requirements from the multiple mobile users. Through minimizing the overall system cost, the optimal resource allocation policy is derived by using the Value Iteration Algorithm. Simulation results are provided, demonstrating the efficiency of the proposed Cloud Broker design.
Highlights
1.1 Context and MotivationCloud computing has become a widely used computing infrastructure based on which many existing cloud vendors
We have proposed a novel an Semi-Markov Decision Process (SMDP)-based formulation of the cloud Broker problem in an Mobile cloud computing (MCC) environment considering the minimization of the overall system cost as target objective
The results show that the Cloud Service Providers (CSP) can ensure a better QoS and gain a higher revenue when they cooperate
Summary
43 4.4 The structure of the optimal policy for service class 2 when p12 = p22= 0.6 MU. 43 4.5 The structure of the optimal policy for service class 1 when p12 = p22= 0.4 MU, 4.6 The structure of the optimal policy for service class 2 when p12 = p22 = 0.4 MU, 4.7 The structure of the optimal policy for Service Class 1 when p12 = p22= 0.4 MU, 4.8 The structure of the optimal policy for Service Class 2 when p12 = p22= 0.4 MU, ix.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have