Dynamic Parallelization of Grid-Enabled Web Services

In a grid environment, it is of primary concern to make efficient use of the resources that are available at run-time. If new computational resources become available, then requests shall also be sent to these newly added resources in order to balance the overall load in the system. However, scheduling of requests in a service grid considers each single service invocation in isolation and determines the most appropriate provider, according to some heuristics. Even when several providers offer the same service, only one of them is chosen. In this paper, we provide a novel approach to the parallelization of individual service requests. This approach makes dynamic use of a set of service providers available at the time the request is being issued. A dynamic service uses meta information on the currently available service providers and their capabilities and splits the original request up into a set of simpler requests of the same service types, submits these requests in parallel to as many service providers as possible, and finally integrates the individual results to the result of the original service request.

[1]  Jiawei Han,et al.  Data Mining: Concepts and Techniques , 2000 .

[2]  Heiko Schuldt,et al.  Data Stream Management and Digital Library Processes on Top of a Hyperdatabase and Grid Infrastructure , 2004, DELOS.

[3]  Ian Foster,et al.  The Grid 2 - Blueprint for a New Computing Infrastructure, Second Edition , 1998, The Grid 2, 2nd Edition.

[4]  Ian T. Foster.  Automatic Generation of Self-Scheduling Programs , 1991, IEEE Trans. Parallel Distributed Syst..

[5]  Hans-Peter Kriegel,et al.  Subspace selection for clustering high-dimensional data , 2004, Fourth IEEE International Conference on Data Mining (ICDM'04).

[6]  Ian T. Foster,et al.  A problem-specific fault-tolerance mechanism for asynchronous, distributed systems , 2000, Proceedings 2000 International Conference on Parallel Processing.

[7]  Luc Moreau,et al.  Towards a Protocol for the Attachment of Semantic Descriptions to Grid Services , 2004, European Across Grids Conference.

[8]  Christian Böhm,et al.  Modelling of classification rules on metabolic patterns including machine learning and expert knowledge , 2005, J. Biomed. Informatics.

[9]  Henri E. Bal,et al.  Efficient load balancing for wide-area divide-and-conquer applications , 2001, PPoPP '01.

[10]  Tim Brecht,et al.  Ajents: towards an environment for parallel, distributed and mobile Java applications , 2000 .

[11]  Heiko Schuldt,et al.  Scalable peer-to-peer process management - the OSIRIS approach , 2004 .

[12]  Jeff T. Linderoth,et al.  Solving large quadratic assignment problems on computational grids , 2002, Math. Program..

[13]  Jeff T. Linderoth,et al.  An enabling framework for master-worker applications on the Computational Grid , 2000, Proceedings the Ninth International Symposium on High-Performance Distributed Computing.

[14]  Francine Berman,et al.  Adaptive scheduling of master/worker applications on distributed computational resources , 2001 .