Computational Technology Resources - CCP

Keywords: service-oriented computing, continuations, scheduling.

Summary

In Service-Oriented Computing (SOC) [3] computational resources are abstracted in terms of services, i.e. computational capabilities not subject to a centralized control exposed to the network through a set of well-defined interfaces and standard protocols.

In the present paper, a service request is fulfilled when the service consumer and provider both agree upon the Quality-of-Service requirements that have to be met [1]. A cost parameter is associated to a service request accounting for the QoS required upon service submission, and the provider receiving the request executes the service with a priority that is a function of this cost to fulfil the consumer request.

In order to offer agreed QoS, and to be able to change quality levels when necessary, service providers need to have control on the execution of their services. In such a way they can accommodate for changing conditions, and they can decide at run-time what Quality-of-Service to supply.

This work presents an infrastructure to model service providers able to control the execution of the services they provide through service suspension and resuming. The infrastructure relies on a continuation programming mechanism [2] to save and restore the service execution state. In this way it is possible to dynamically control the execution of services and to change at run-time parameters affecting service provision either driven by consumer or system requirements.

In order to test the flexibility of the infrastructure, a cost-based resource-sharing scheduler [4] has been implemented. The paper reports tests of simulations of two policies implemented within the scheduler of the proposed infrastructure.

As a general result, the two cost-based resource-sharing policies showed two different types of behaviour with respect to the system workload (i.e. the number of requests). While one policy better meets user requirements if the services are requested with higher costs although inhibiting cpu access (unfairness) to low cost services. The other policy tries to meet high priority requests and at the same time it guarantees minimal cpu utilization to low priority requests during all the scheduling epochs. The policy to apply depends on the requirements (system and user-defined) to be met.

References

1: J. MacLaren, R. Sakellariou, J. Garibaldi, D. Ouelhadj, "Towards service level agreement based scheduling on the grid", Proc. of 2nd European Across Grids Conference, 2004.
2: C.T. Haynes, D.P. Friedman, E.E. Kohlbecker, "Program Transformation and Programming Environments", Programming with Continuations, Springer-Verlag, 1984.
3: I. Foster, C. Kesselman, J. Nick, S. Tuecke, "The physiology of the grid: An open grid service architecture for distributed system integration", Tech. report Open Grid Service Infrastructure WG, 2002.
4: T. Newhouse, J. Pasquale, "A user-level framework for scheduling within service execution environments", Proc. of 2004 IEEE Intern. Conf. on Services Computing, IEEE Computer Society, 2004. doi:10.1109/SCC.2004.1358020

purchase the full-text of this paper (price £20)

go to the previous paper
go to the next paper
return to the table of contents
return to the book description
purchase this book (price £72 +P&P)

	Computational & Technology Resources an online resource for computational, engineering & technology publications
	not logged in - login
Front Page Browse CCP CSETS CTR IJRT Other Authors Search Purchase Guide FAQ Contact us	Civil-Comp Proceedings ISSN 1759-3433 CCP: 90 PROCEEDINGS OF THE FIRST INTERNATIONAL CONFERENCE ON PARALLEL, DISTRIBUTED AND GRID COMPUTING FOR ENGINEERING Edited by: Paper 29 A Continuation Based Middleware to Experiment with Economy Based Service Scheduling C. Di Napoli and M. Giordano C.N.R. - Istituto di Cibernetica "E. Caianiello", Pozzuoli, Naples, Italy doi:10.4203/ccp.90.29 purchase the full-text of this paper Full Bibliographic Reference for this paper C. Di Napoli, M. Giordano, "A Continuation Based Middleware to Experiment with Economy Based Service Scheduling", in , (Editors), "Proceedings of the First International Conference on Parallel, Distributed and Grid Computing for Engineering", Civil-Comp Press, Stirlingshire, UK, Paper 29, 2009. doi:10.4203/ccp.90.29 Keywords: service-oriented computing, continuations, scheduling. Summary In Service-Oriented Computing (SOC) [3] computational resources are abstracted in terms of services, i.e. computational capabilities not subject to a centralized control exposed to the network through a set of well-defined interfaces and standard protocols. In the present paper, a service request is fulfilled when the service consumer and provider both agree upon the Quality-of-Service requirements that have to be met [1]. A cost parameter is associated to a service request accounting for the QoS required upon service submission, and the provider receiving the request executes the service with a priority that is a function of this cost to fulfil the consumer request. In order to offer agreed QoS, and to be able to change quality levels when necessary, service providers need to have control on the execution of their services. In such a way they can accommodate for changing conditions, and they can decide at run-time what Quality-of-Service to supply. This work presents an infrastructure to model service providers able to control the execution of the services they provide through service suspension and resuming. The infrastructure relies on a continuation programming mechanism [2] to save and restore the service execution state. In this way it is possible to dynamically control the execution of services and to change at run-time parameters affecting service provision either driven by consumer or system requirements. In order to test the flexibility of the infrastructure, a cost-based resource-sharing scheduler [4] has been implemented. The paper reports tests of simulations of two policies implemented within the scheduler of the proposed infrastructure. As a general result, the two cost-based resource-sharing policies showed two different types of behaviour with respect to the system workload (i.e. the number of requests). While one policy better meets user requirements if the services are requested with higher costs although inhibiting cpu access (unfairness) to low cost services. The other policy tries to meet high priority requests and at the same time it guarantees minimal cpu utilization to low priority requests during all the scheduling epochs. The policy to apply depends on the requirements (system and user-defined) to be met. References 1 J. MacLaren, R. Sakellariou, J. Garibaldi, D. Ouelhadj, "Towards service level agreement based scheduling on the grid", Proc. of 2nd European Across Grids Conference, 2004. 2 C.T. Haynes, D.P. Friedman, E.E. Kohlbecker, "Program Transformation and Programming Environments", Programming with Continuations, Springer-Verlag, 1984. 3 I. Foster, C. Kesselman, J. Nick, S. Tuecke, "The physiology of the grid: An open grid service architecture for distributed system integration", Tech. report Open Grid Service Infrastructure WG, 2002. 4 T. Newhouse, J. Pasquale, "A user-level framework for scheduling within service execution environments", Proc. of 2004 IEEE Intern. Conf. on Services Computing, IEEE Computer Society, 2004. doi:10.1109/SCC.2004.1358020 purchase the full-text of this paper (price £20) go to the previous paper go to the next paper return to the table of contents return to the book description purchase this book (price £72 +P&P)
Back to top	©Civil-Comp Limited 2023 - terms & conditions