
This Article  
 
Share  
Bibliographic References  
Add to:  
Digg Furl Spurl Blink Simpy Del.icio.us Y!MyWeb  
Search  
 
ASCII Text  x  
Ligang He, Stephen A. Jarvis, Daniel P. Spooner, Hong Jiang, Donna N. Dillenberger, Graham R. Nudd, "Allocating NonRealTime and Soft RealTime Jobs in Multiclusters," IEEE Transactions on Parallel and Distributed Systems, vol. 17, no. 2, pp. 99112, February, 2006.  
BibTex  x  
@article{ 10.1109/TPDS.2006.18, author = {Ligang He and Stephen A. Jarvis and Daniel P. Spooner and Hong Jiang and Donna N. Dillenberger and Graham R. Nudd}, title = {Allocating NonRealTime and Soft RealTime Jobs in Multiclusters}, journal ={IEEE Transactions on Parallel and Distributed Systems}, volume = {17}, number = {2}, issn = {10459219}, year = {2006}, pages = {99112}, doi = {http://doi.ieeecomputersociety.org/10.1109/TPDS.2006.18}, publisher = {IEEE Computer Society}, address = {Los Alamitos, CA, USA}, }  
RefWorks Procite/RefMan/Endnote  x  
TY  JOUR JO  IEEE Transactions on Parallel and Distributed Systems TI  Allocating NonRealTime and Soft RealTime Jobs in Multiclusters IS  2 SN  10459219 SP99 EP112 EPD  99112 A1  Ligang He, A1  Stephen A. Jarvis, A1  Daniel P. Spooner, A1  Hong Jiang, A1  Donna N. Dillenberger, A1  Graham R. Nudd, PY  2006 KW  Scheduling KW  parallel systems KW  distributed systems KW  realtime systems KW  numerical algorithms. VL  17 JA  IEEE Transactions on Parallel and Distributed Systems ER   
Abstract—This paper addresses workload allocation techniques for two types of sequential jobs that might be found in multicluster systems, namely, nonrealtime jobs and soft realtime jobs. Two workload allocation strategies, the Optimized mean Response Time (ORT) and the Optimized mean Miss Rate (OMR), are developed by establishing and numerically solving two optimization equation sets. The ORT strategy achieves an optimized mean response time for nonrealtime jobs, while the OMR strategy obtains an optimized mean miss rate for soft realtime jobs over multiple clusters. Both strategies take into account average system behaviors (such as the mean arrival rate of jobs) in calculating the workload proportions for individual clusters and the workload allocation is updated dynamically when the change in the mean arrival rate reaches a certain threshold. The effectiveness of both strategies is demonstrated through theoretical analysis. These strategies are also evaluated through extensive experimental studies and the results show that when compared with traditional strategies, the proposed workload allocation schemes significantly improve the performance of job scheduling in multiclusters, both in terms of the mean response time (for nonrealtime jobs) and the mean miss rate (for soft realtime jobs).
[1] B. Adelberg, H. GarciaMolina, and B. Kao, “Emulating Soft Realtime Scheduling Using Traditional Operating System Schedulers,” Proc. 1994 IEEE RealTime Systems Symp., pp. 292298, 1994.
[2] O. Aumage, “Heterogeneous MultiCluster Networking with the Madeleine III Communication Library,” Proc. 16th Int'l Parallel and Distributed Processing Symp. (IPDPS 2002), pp. 172183, 2002.
[3] S.A. Banawan and N.M. Zeidat, “A Comparative Study of Load Sharing in Heterogeneous Multicomputer Systems,” Proc. 25th Ann. Simulation Symp., pp. 2231, 1992.
[4] S. Banen, A.I.D. Bucur, and D.H.J. Epema, “A MeasurementBased Simulation Study of Processor CoAllocation in Multicluster Systems,” Proc. Ninth Workshop Job Scheduling Strategies for Parallel Processing, pp. 105128, 2003.
[5] M. Barreto, R. Avila, and P. Navaux, “The MultiCluster Model to the Integrated Use of Multiple Workstation Clusters,” Proc. Third Workshop Personal ComputerBased Networks of Workstations, pp. 7180, 2000.
[6] D.P. Bertsekas, Constrained Optimization and Lagrange Multiplier Methods, ISBN: 1886529043, 1996.
[7] A. Bivens, D. Dillenberger, C. Chhuor, G. Ferris, W. Chou, and J. Fenton, “CISCO Load Balancing with SASP and Enterprise Workload Management,” Oct. 2004.
[8] G. Bolch, S. Greiner, H. de Meer, and K.S. Trivedi, Queueing Networks and Markov Chains, Modeling and Performance Evaluation with Computer Science Applications. John Wiley & Sons, 1998.
[9] A.I.D. Bucur and D.H.J. Epema, “The Maximal Utilization of Processor CoAllocation in Multicluster Systems,” Proc. Int'l Parallel and Distributed Processing Symp. (IPDPS 2003), pp. 6069, 2003.
[10] K. Chen and L. Decreusefond, “Just How Bad is the FIFO Discipline for Handling Randomly Arriving TimeCritical Messages,” Proc. 1995 IEEE Int'l Workshop Factory Comm. Systems, pp. 183190, 1995.
[11] Y.C. Chow and W.H. Kohler, “Models for Dynamic Load Balancing in Heterogeneous Multiple Processor System,” IEEE Trans. Computers, vol. 28, no. 5, pp. 354361, 1979.
[12] M. Colajanni and P.S. Yu, “Dynamic Load Balancing in Geographically Distributed Heterogeneous Web Servers,” Proc. 18th Int'l Conf. Distributed Computing Systems, pp. 295302, 1998.
[13] M. Chu, K. Fan, and S. Mahlke, “RegionBased Hierarchical Operation Partitioning for Multicluster Processors,” Proc. ACM SIGPLAN 2003 Conf. Programming Language Design and Implementation, pp. 300311, 2003.
[14] M.M. Deris, J.H. Abawajy, and H.M. Suzuri, “An Efficient Replicated Data Access Approach for LargeScale Distributed Systems,” Proc. Fourth IEEE Int'l Symp. Cluster Computing and the Grid, pp. 588594, 2004.
[15] L. He, S.A. Jarvis, D.P. Spooner, X. Chen, and G.R. Nudd, “Hybrid PerformanceBased Workload Management for Multiclusters and Grids,” IEE Proc. Software, special issue on performance eng., vol. 151, no. 5, pp. 224231, Oct. 2004.
[16] L. He, S.A. Jarvis, D.P. Spooner, X. Chen, and G.R. Nudd, “Dynamic Scheduling of Parallel Jobs with QoS Demands in Multiclusters and Grids,” Proc. Fifth IEEE/ACM Int'l Workshop Grid Computing (Grid 2004), pp. 402409, 2004.
[17] L. He, S.A. Jarvis, D.P. Spooner, and G.R. Nudd, “Optimising Static Workload Allocation in Multiclusters,” Proc. 18th IEEE Int'l Parallel and Distributed Processing Symp. (IPDPS '04), 2004.
[18] L. He, S.A. Jarvis, D.P. Spooner, and G.R. Nudd, “Dynamic Scheduling of Parallel RealTime Jobs by Modelling Spare Capabilities in Heterogeneous Clusters,” Proc. Fifth IEEE Int'l Conf. Cluster Computing (Cluster '03), pp. 210, 2003.
[19] L. He, Z. Han, H. Jin, and L. Pang, “DAGBased Parallel Real Time Task Scheduling Algorithm on a Cluster,” Proc. Seventh Int'l Conf. Parallel and Distributed Processing Techniques and Applications (PDPTA 2000), pp. 437443, 2000.
[20] B. Kao and H. GarciaMolina, “Scheduling Soft RealTime Jobs over Dual NonRealTime Servers,” IEEE Trans. Parallel and Distributed Systems, vol. 7, no. 1, pp. 5668, Jan. 1996.
[21] L. Kleinrock, Queueing System. John Wiley & Sons, 1975.
[22] R. Leslie and S. McKenzie, “Evaluation of Load Sharing Algorithms for Heterogeneous Distributed Systems,” Computer Comm., vol. 22, no. 4, pp. 376389, 1999.
[23] Z. Liu, M.S. Squillante, and J.L. Wolf, “On Maximizing ServiceLevelAgreement Profits,” Proc. Third ACM Conf. Electronic Commerce, pp. 213223, 2001.
[24] N.G. Shivaratri, P. Krueger, and M. Singhal, “Load Distribution for Locally Distributed Systems,” Computer, vol. 8, no. 12, pp. 3344, Dec. 1992.
[25] X.Y. Tang and S.T. Chanson, “Optimizing Static Job Scheduling in a Network of Heterogeneous Computers,” Proc. 29th Int'l Conf. Parallel Processing, pp. 373382, 2000.
[26] M.Q. Xu, “Effective Metacomputing Using LSF Multicluster,” Proc. First Int'l Symp. Cluster Computing and the Grid (CCGrid '01), pp. 100105, 2001.
[27] S. Zhou, “A TraceDriven Simulation Study of Dynamic Load Balancing,” IEEE Trans. Software Eng., vol. 14, no. 9, pp. 13271341, 1988.
[28] W. Zhu and B. Fleisch, “Performance Evaluation of Soft RealTime Scheduling on a Multicomputer Cluster,” Proc. 20th Int'l Conf. Distributed Computing Systems (ICDCS 2000), pp. 610617, 2000.