Resource scheduling with conflicting objectives in grid environments: Model and evaluation

The paper is to consider resource scheduling with conflicting objectives in the grid environment. The objectives of the grid users, the grid resources and the grid system clash with each other. Grid users want to access enough system resources to achieve the desired level of quality of service (QoS). Resource providers pay more attention to the performance of their resources. Our resource scheduling employs market strategies to determine which jobs are executed at what time on which resources and at what prices. A grid resource provider uses its utility function to maximize its profit and a grid user uses its utility function to complete tasks while minimizing its spending. The paper proposes grid system objective optimization scheduling that provides a joint optimization of objectives for both the resource provider and grid user, which combines the benefits of both resource provider objective optimization and user objective optimization. Experiments are designed to study the performances of three resource-scheduling optimization algorithms. Performance metrics are classified into efficiency metrics, utility metrics and time metrics.

[1]  Layuan Li,et al.  Competitive proportional resource allocation policy for computational grid , 2004, Future Gener. Comput. Syst..

[2]  Layuan Li,et al.  The use of economic agents under price driven mechanism in grid resource management , 2004, J. Syst. Archit..

[3]  Jennifer Healey,et al.  QoS-Constrained Resource Allocation for a Grid-Based Multiple Source Electrocardiogram Application , 2004, ICCSA.

[4]  Ian T. Foster,et al.  End-to-end quality of service for high-end applications , 2004, Comput. Commun..

[5]  Layuan Li,et al.  Multi economic agent interaction for optimizing the aggregate utility of grid users in computational grid , 2006, Appl. Intell..

[6]  Layuan Li,et al.  A distributed utility-based two level market solution for optimal resource scheduling in computational grid , 2005, Parallel Comput..

[7]  Li Chunlin,et al.  Apply agent to build grid service management , 2003 .

[8]  Rajkumar Buyya,et al.  A Deadline and Budget Constrained Cost-Time Optimisation Algorithm for Scheduling Task Farming Applications on Global Grids , 2002, ArXiv.

[9]  John P. Lehoczky,et al.  Scalable resource allocation for multi-processor QoS optimization , 2003, 23rd International Conference on Distributed Computing Systems, 2003. Proceedings..

[10]  David Abramson,et al.  A case for economy grid architecture for service oriented grid computing , 2001, Proceedings 15th International Parallel and Distributed Processing Symposium. IPDPS 2001.

[11]  Atakan Dogan,et al.  A comparison of static QoS-based scheduling heuristics for a meta-task with multiple QoS dimensions in heterogeneous computing , 2004, 18th International Parallel and Distributed Processing Symposium, 2004. Proceedings..

[12]  Li Chunlin,et al.  Agent framework to support the computational grid , 2004 .

[13]  Omer F. Rana,et al.  Supporting QoS-based discovery in service-oriented Grids , 2003, Proceedings International Parallel and Distributed Processing Symposium.

[14]  Atakan Dogan,et al.  Scheduling Independent Tasks with QoS Requirements in Grid Computing with Time-Varying Resource Prices , 2002, GRID.

[15]  Daniel P. Siewiorek,et al.  On quality of service optimization with discrete QoS options , 1999, Proceedings of the Fifth IEEE Real-Time Technology and Applications Symposium.

[16]  John P. Lehoczky,et al.  Integrated resource management and scheduling with multi-resource constraints , 2004, 25th IEEE International Real-Time Systems Symposium.

[17]  Ramin Yahyapour,et al.  Economic Scheduling in Grid Computing , 2002, JSSPP.

[18]  Sanjay Jha,et al.  G-QoSM: Grid Service Discovery Using QoS Properties , 2002, Comput. Artif. Intell..