Abstract
In systems consistingof multiple clusters of processors interconnected by relatively slow connections such as our Distributed ASCI1 Supercomputer (DAS), jobs may request co-allocation, i.e., the simultaneous allocation of processors in different clusters. The performance of co-allocation may be severely impacted by the slowintercluster connections, and by the types of job requests.We distinguish different job request types ranging from ordered requests that specify the numbers of processors needed in each of the clusters, to flexible requests that only specify a total. We simulate multicluster systems with the FCFS policy — and with two policies for placinga flexible request, one tries to balance cluster loads and one tries to fill clusters completely—to determine the response times under workloads consistingof a single or of different request types for different communication speeds across the intercluster connections. In addition to a synthetic workload, we also consider a workload derived from measurements of a real application on the DAS. We find that the communication speed difference has a severe impact on response times, that a relatively small amount of capacity is lost due to communication, and that for a mix of request types, the performance is determined not only by the separate behaviours of the different types of requests, but also by the way in which they interact.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
References
K. Aida, H. Kasahara and S. Narita. Job SchedulingScheme for Pure Space SharingAmong Rigid Jobs. In Job Scheduling Strategies for Parallel Processing, Lecture Notes in Computer Science 1459, pages 98–121. Springer-Verlag, 1998.
A.H. Alhusaini, V.K. Prasanna, and C.S. Raghavendra, A Framework for Mapping with Resource Co-Allocation in Heterogeneous Computing Systems, Proc. 9th Heterogeneous ComputingW orkshop (HCW2000), C.S. Raghavendra (ed.), pp. 273–286, 2000.
Bal, H.E., Plaat, A., Bakker, M.G., Dozy, P., Hofman, R.F.H.: Optimizing Parallel Applications forWide-Area Clusters. Proceedings of the 12th International Parallel Processing Symposium (1998) 784–790
Brecht, T.B.: An Experimental Evaluation of Processor Pool-Based Schedulingfor Shared-Memory NUMA multiprocessors. Job Scheduling Strategies for Parallel Processing, Lecture Notes in Computer Science 1291 Springer-Verlag (1997) 139–165
Bal, H.E., et al.: The Distributed ASCI Supercomputer Project. ACM OperatingSystems Review 34(4) (2000) 76–96
Feitelson, D.G., Rudolph, L.: Toward Convergence in Job Schedulers for Parallel Supercomputers. Job Scheduling Strategies for Parallel Processing, Lecture Notes in Computer Science 1162 Springer-Verlag (1996) 1–26
Feitelson, D.G., Rudolph, L.: Theory and Practice in Parallel Job SchedulingJob Scheduling Strategies for Parallel Processing, Lecture Notes in Computer Science 1291 Springer-Verlag (1997) 1–34
Feitelson, D.G., Jette, M.A.: Improved Utilization and Responsiveness with GangScheduling. Job Scheduling Strategies for Parallel Processing, Lecture Notes in Computer Science 1291 Springer-Verlag (1997) 238–261
Feitelson, D.G.: PackingSchemes for GangScheduling. Job SchedulingStrategies for Parallel Processing, Lecture Notes in Computer Science 1162 Springer-Verlag (1996) 89–110
Patton Jones, J., Nitzberg, B.: Scheduling for Parallel Supercomputing: A Historical Perspective of Achievable Utilization. Job Scheduling Strategies for Parallel Processing, Lecture Notes in Computer Science 1659 Springer-Verlag (1999) 1–16
Kielmann, T., Hofman, R.F.H., Bal, H.E., Plaat, A., Bhoedjang, R.A.F.: MagPIe: MPI’s Collective Communication Operations for ClusteredWide Area Systems. ACMSIGPLAN Symposium on Principles and Practice of Parallel Programming (1999) 131–140
Leinberger, W., Karypis, G., Kumar, V.: Milti-Capacity Bin Packing Algorithms with Applications to Job Schedulingunder Multiple Constraints. Proc. 1999 Int’l Conference on Parallel Processing(1999) 404–412
Foster, I., Kesselman, C. (eds): The Grid: Blueprint for a New ComputingInfrastructure. Morgan Kaufmann (1999)
Snell, Q., Clement, M., Jackson, D., Gregory, C.: The Performance Impact of Advance Reservation Meta-Scheduling. Job Scheduling Strategies for Parallel Processing, Lecture Notes in Computer Science 1911 Springer-Verlag (2000) 137–153
Bucur, A.I.D., Epema, D.H.J: The Influence of the Structure and Sizes of Jobs on the Performance of Co-Allocation. Job Scheduling Strategies for Parallel Processing, Lecture Notes in Computer Science 1911 Springer-Verlag (2000) 154–173
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2001 Springer-VerlagBerlin Heidelberg
About this paper
Cite this paper
Bucur, A., Epema, D. (2001). The Influence of Communication on the Performance of Co-allocation. In: Feitelson, D.G., Rudolph, L. (eds) Job Scheduling Strategies for Parallel Processing. JSSPP 2001. Lecture Notes in Computer Science, vol 2221. Springer, Berlin, Heidelberg. https://doi.org/10.1007/3-540-45540-X_5
Download citation
DOI: https://doi.org/10.1007/3-540-45540-X_5
Published:
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-42817-6
Online ISBN: 978-3-540-45540-0
eBook Packages: Springer Book Archive