The Performance of Processor Co-Allocation in Multicluster Systems

Authors:
A. I. D. Bucur;D. H. J. Epema
Affiliations:
-;-
Venue:
CCGRID '03 Proceedings of the 3st International Symposium on Cluster Computing and the Grid
Year:
2003

Citing 0
Cited 19

Characterization of Bandwidth-Aware Meta-Schedulers for Co-Allocating Jobs Across Multiple Clusters

The Journal of Supercomputing
The Impact of Information Availability and Workload Characteristics on the Performance of Job Co-allocation in Multi-clusters

ICPADS '06 Proceedings of the 12th International Conference on Parallel and Distributed Systems - Volume 1
Ensuring Fairness Among Participating Clusters During Multi-site Parallel Job Scheduling

ICPADS '06 Proceedings of the 12th International Conference on Parallel and Distributed Systems - Volume 2
Using checkpointing to recover from poor multi-site parallel job scheduling decisions

Proceedings of the 5th international workshop on Middleware for grid computing: held at the ACM/IFIP/USENIX 8th International Middleware Conference
VARQ: virtual advance reservations for queues

HPDC '08 Proceedings of the 17th international symposium on High performance distributed computing
A two-class parallel system with general service times of the parallel class

Journal of Computer and System Sciences
Resource Matching in Non-dedicated Multicluster Environments

High Performance Computing for Computational Science - VECPAR 2008
Job co-allocation strategies for multiple high performance computing clusters

Cluster Computing
The impact of error in user-provided bandwidth estimates on multi-site parallel job scheduling performance

PDCS '07 Proceedings of the 19th IASTED International Conference on Parallel and Distributed Computing and Systems
On grid performance evaluation using synthetic workloads

JSSPP'06 Proceedings of the 12th international conference on Job scheduling strategies for parallel processing
Towards feasible and effective load sharing in a heterogeneous computational grid

GPC'07 Proceedings of the 2nd international conference on Advances in grid and pervasive computing
Using moldability to improve scheduling performance of parallel jobs on computational grid

GPC'08 Proceedings of the 3rd international conference on Advances in grid and pervasive computing
Group-wise performance evaluation of processor co-allocation in multi-cluster systems

JSSPP'07 Proceedings of the 13th international conference on Job scheduling strategies for parallel processing
Deadline-guarantee-enhanced co-allocation for parameter sweep application in grid

ICC'09 Proceedings of the 2009 IEEE international conference on Communications
Multisite co-allocation algorithms for computational grid

IPDPS'06 Proceedings of the 20th international conference on Parallel and distributed processing
A dynamic co-allocation service in multicluster systems

JSSPP'04 Proceedings of the 10th international conference on Job Scheduling Strategies for Parallel Processing
An M/M/2 parallel system model with pure space sharing among rigid jobs

Mathematical and Computer Modelling: An International Journal
Improving grid performance through processor allocation considering both speed heterogeneity and resource fragmentation

The Journal of Supercomputing
Large improvements in application throughput of long-running multi-component applications using batch grids

Concurrency and Computation: Practice & Experience

Quantified Score

Hi-index	0.00

Visualization

Abstract

In systems consisting of multiple clusters of processorswhich are interconnected by relatively slow communicationlinks and which employ space sharing for scheduling jobs,such as our Distributed ASCI 1 Supercomputer (DAS), co-allocation,i.e., the simultaneous allocation of processorsto single jobs in different clusters, may be required. Westudy the performance of co-allocation by means of simulationsfor the mean response time of jobs depending on thestructure and sizes of jobs, the scheduling policy, and thecommunication speed ratio. Our main conclusion is thatfor current communication speed ratios in multiclusters, co-allocationis a viable option.