Farag Azzedin

Integrating trust into grid resource management systems

Grid computing systems that have been the focus of much research in recent years provide a virtual framework for controlled sharing of resources across institutional boundaries. Security is a major concern in any system that enables remote execution. Several techniques can be used for providing security in grid systems including sandboxing, encryption, and other access control and authentication mechanisms. The additional overhead caused by these mechanisms may negate the performance advantages gained by grid computing. Hence, we contend that it is essential for the scheduler to consider the security implications while performing resource allocations. In this paper, we present a trust model for grid systems and show how the model can be used to incorporate security implications into scheduling algorithms. Three scheduling heuristics that can be used in a grid system are modified to incorporate the trust notion and simulations are performed to evaluate the performance.

A Synchronous Co-Allocation Mechanism for Grid Computing Systems

Grid computing systems are emerging as a computing infrastructure that will enable the use of wide-area network computing systems for a variety of challenging applications. One of these is the ever increasing demand for multimedia from users engaging in a wide range of activities such as scientific research, education, commerce, and entertainment. To provide an adequate level of service to multimedia applications, it is often necessary to simultaneously allocate resources including predetermined capacities from interconnecting networks to the applications. The simultaneous allocation of resources is often referred to as co-allocation in the Grid literature. In this paper, we formally define the co-allocation problem and propose a novel scheme called synchronous queuing (SQ) for implementing co-allocation with quality of service (QoS) assurances in Grids. Unlike existing approaches, SQ does not require advance reservation capabilities at the resources. This enables an SQ-based approach to over subscribe the resources and hence improve resource utilization. The simulation studies performed to evaluate SQ indicate that it outperforms an QoS-based scheme with strict admission control by a significant margin.

Mobile interfaces to computational, data, and service grid systems

This article briefly describes the issues related to providing mobile access to computational, data, and service Grids. Then it describes our preliminary efforts to enhance computational and service Grids to handle mobile access. In particular, we focus on how the HARNESS mobile extensions project approaches the problem of enhancing Grid computers with mobile access and we describe how the InviNet project provides access to service Grids with consistent quality of service.

Trust modeling for peer-to-peer based computing systems

The peer-to-peer approach to design large-scale systems has significant benefits including scalability, low cost of ownership, robustness, and ability to provide site autonomy. However, this approach has several drawbacks as well including trust issues and lack of coordination and control among the peers. We present a trust model for a peer-to-peer structured large-scale network computing system and completely define the trust model and describe the schemes used in it. Central to the model is the idea of maintaining a recommender network that can be used to obtain references about a target domain. Simulation results indicate that the trust model is capable of building and maintaining trust and also identifying the bad domains.

MetaGrid: a scalable framework for wide-area service deployment and management

Presents an architecture called the MetaGrid based on Grid computing concepts for resource provisioning for wide-area network-enabled applications. Resource provisioning for wide-area applications can involve coordinated allocation of computing and communication resources. A Grid computing system provides a virtual framework that facilitates controlled resource sharing among different institutions. The MetaGrid extends the Grid computing systems in two major ways: (a) introduces a notion of SubGrid that provides a coarse-grained resource allocation class and (b) introduces a framework for interconnecting Grids by facilitating peering, trading, and brokering among the different Grids. The paper presents (a) the overall architecture of the MetaGrid with a description of the different functional components, (b) the resource allocation model that is introduced by the notion of SubGrids, and (c) strategies of interconnecting Grids.