|
| This Article | ||
| ||
| Share | ||
| Bibliographic References | ||
| Add to: | ||
 Digg  Furl  Spurl  Blink  Simpy  Google  Del.icio.us  Y!MyWeb | ||
| Search | ||
| ||
| ASCII Text | x | ||
| M.H. Willebeek-LeMair, A.P. Reeves, "Strategies for Dynamic Load Balancing on Highly Parallel Computers," IEEE Transactions on Parallel and Distributed Systems, vol. 4, no. 9, pp. 979-993, September, 1993. | |||
| BibTex | x | ||
| @article{ 10.1109/71.243526, author = {M.H. Willebeek-LeMair and A.P. Reeves}, title = {Strategies for Dynamic Load Balancing on Highly Parallel Computers}, journal ={IEEE Transactions on Parallel and Distributed Systems}, volume = {4}, number = {9}, issn = {1045-9219}, year = {1993}, pages = {979-993}, doi = {http://doi.ieeecomputersociety.org/10.1109/71.243526}, 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 - Strategies for Dynamic Load Balancing on Highly Parallel Computers IS - 9 SN - 1045-9219 SP979 EP993 EPD - 979-993 A1 - M.H. Willebeek-LeMair, A1 - A.P. Reeves, PY - 1993 KW - Index Termsdynamic load balancing; highly parallel computers; multicomputer systems; load balancing communication overhead; Inter iPSC/2 hypercube; parallel processing; performance evaluation; resource allocation; synchronisation VL - 4 JA - IEEE Transactions on Parallel and Distributed Systems ER - | |||
Dynamic load balancing strategies for minimizing the execution time of single applications running in parallel on multicomputer systems are discussed. Dynamic load balancing (DLB) is essential for the efficient use of highly parallel systems when solving non-uniform problems with unpredictable load estimates. With the evolution of more highly parallel systems, centralized DLB approaches which make use of a high degree of knowledge become less feasible due to the load balancing communication overhead. Five DLB strategies are presented which illustrate the tradeoff between 1) knowledge - the accuracy of each balancing decision, and 2) overhead - the amount of added processing and communication incurred by the balancing process. All five strategies have beenimplemented on an Inter iPSC/2 hypercube.
[1] M. Willebeek-LeMair and A. P. Reeves, "Region growing on a hypercube multiprocessor," inProc. 3rd Conf. Hypercube Concurrent Comput. and Appl., 1988, pp. 1033-1042.
[2] T. L. Casavant and J. G. Kuhl, "A taxonomy of scheduling in general-purpose distributed computing systems,"IEEE Trans. Software Eng., vol. 14, no. 2, pp. 141-154, Feb. 1988.
[3] Y.-T. Wang and R. J. T. Morris, "Load sharing in distributed systems,"IEEE Trans. Comput., vol. C-34, pp. 204-217, Mar. 1985.
[4] M. Berger and S. H. Bokhari, "A partitioning strategy for nonuniform problems on multiprocessors,"IEEE Trans. Comput., vol. C-36, pp. 570-580, May 1987.
[5] G. C. Fox, "A review of automatic load balancing and decomposition methods for the hypercube," California Institute of Technology, C3P- 385, Nov. 1986.
[6] K. Ramamritham, J. A. Stankovic, and W. Zhao, "Distributed scheduling of tasks with deadlines and resource requirements,"IEEE Trans. Comput., pp. 1110-1123, Aug. 1989.
[7] K. M. Baumgartner, R. M. Kling, and B. W. Wah, "Implementation of GAMMON: An efficient load balancing strategy for a local computer system," inProc. 1989 Int. Conf. Parallel Processing, vol. 2, Aug. 1989, pp. 77-80.
[8] F. C. H. Lin and R. M. Keller, "The gradient model load balancing method,"IEEE Trans. Software Eng., vol. SE-13, pp. 32-38, Jan. 1987.
[9] L. V. Kale, "Comparing the performance of two dynamic load distribution methods," inProc. 1988 Int. Conf. Parallel Processing, Aug. 1988, pp. 8-12.
[10] J. Hong, X. Tan, and M. Chen, "From local to global: An analysis of nearest neighbor balancing on hypercube," inProc. Third Conf. Hypercube Concurrent Comput. and Appl., Jan. 1988.
[11] K. G. Shin and Y.-C. Chang, "Load sharing in distributed real-time systems with state-change broadcasts,"IEEE Trans. Comput., pp. 1124-1142, Aug. 1989.
[12] V. A. Saletore, "A distrubuted and adaptive dynamic load balancing scheme for parallel processing of medium-grain tasks," inProc. Fifth Distributed Memory Comput. Conf., Apr. 1990, pp. 995-999.
[13] W. Shu and L. V. Kale, "A dynamic scheduling strategy for the Charekernel system," inProc. ACM Supercomput. Conf., 1989, pp. 389-398.
[14] M. Willebeek-LeMair and A. P. Reeves, "A general dynamic load balancing model for parallel computers," Tech. Rep. EE-CEG-89-1, Cornell School of Electrical Engineering, 1989.
[15] G. Cybenko, "Dynamic load balancing for distributed memory multiprocessors,"J. Parallel and Distributed Comput., vol. 7:279-301, October, 1989.
[16] D. P. Bertsekas and J. N. Tsitsiklis,Parallel and Distributed Computations. Englewood Cliffs, NJ: Prentice-Hall, 1989.
[17] K. M. Dragon and J. L. Gustafson, "A low-cost hypercube load balance algorithm," inProc. Fourth Conf. Hypercubes, Concurrent Comput. and Appl., 1989, pp. 583-590.
[18] M. Hailperin, "Load balancing for massively-parallel soft-real-time systems," inFrontiers '88: 2nd Symp. Frontiers of Massively Parallel Computation. New York: IEEE, 1988, pp. 159-163.
[19] D. L. Eager, E. Lazowska, and J. Zahorjan, "A comparison of receiver-initiated and sender-initiated adaptive loading,"Perform. Eval., vol. 6, 1986.
[20] R. P. Pargas and E. D. Wooster, "Branch-and-bound algorithms on a hypercube," inProc. Third Conf. Hypercube Concurrent Comput. and Appl., Jan. 1988.
[21] V. N. Rao and V. Kumar, "Parallel depth-first search, Part I: Implementation,"Int. J. Parallel Programming, vol. 16, no. 6, pp. 479-499, 1987.
[22] V. Kumar and V. N. Rao, "Parallel depth-first search, Part II: Analysis,"Int. J. Parallel Program., vol. 16, no. 6, pp. 501-519, 1987.