Electronic Thesis and Dissertation Repository


Doctor of Philosophy


Electrical and Computer Engineering


Hamada Ghenniwa


In the Grid/Cloud environment, applications or services and resources belong to different organizations with different objectives. Entities in the Grid/Cloud are autonomous and self-interested; however, they are willing to share their resources and services to achieve their individual and collective goals. In such open environment, the scheduling decision is a challenge given the decentralized nature of the environment. Each entity has specific requirements and objectives that need to achieve. In this thesis, we review the Grid/Cloud computing technologies, environment characteristics and structure and indicate the challenges within the resource scheduling. We capture the Grid/Cloud scheduling model based on the complete requirement of the environment. We further create a mapping between the Grid/Cloud scheduling problem and the combinatorial allocation problem and propose an adequate economic-based optimization model based on the characteristic and the structure nature of the Grid/Cloud. By adequacy, we mean that a comprehensive view of required properties of the Grid/Cloud is captured. We utilize the captured properties and propose a bidding language that is expressive where entities have the ability to specify any set of preferences in the Grid/Cloud and simple as entities have the ability to express structured preferences directly. We propose a winner determination model and mechanism that utilizes the proposed bidding language and finds a scheduling solution. Our proposed approach integrates concepts and principles of mechanism design and classical scheduling theory. Furthermore, we argue that in such open environment privacy concerns by nature is part of the requirement in the Grid/Cloud. Hence, any scheduling decision within the Grid/Cloud computing environment is to incorporate the feasibility of privacy protection of an entity. Each entity has specific requirements in terms of scheduling and privacy preferences. We analyze the privacy problem in the Grid/Cloud computing environment and propose an economic based model and solution architecture that provides a scheduling solution given privacy concerns in the Grid/Cloud. Finally, as a demonstration of the applicability of the approach, we apply our solution by integrating with Globus toolkit (a well adopted tool to enable Grid/Cloud computing environment). We also, created simulation experimental results to capture the economic and time efficiency of the proposed solution.