Abstract
Grid architecture integrates geographically distributed nodes to manage and provide resources to execute scientific applications. For data locality, applications with different computational phases require data redistribution for realignment. The tradeoff between high efficiency computation and communication cost of data redistribution accompanies. This paper introduces a research model and two methods to derive new lists of processor logical id according to the characteristics of heterogeneous network. Both methods provide choices of more low-cost communication schedules in grid. The simulations show both proposed methods yield outstanding performance in grid.
Similar content being viewed by others
References
Beaumont O, Legrand A, Robert Y (2003) Optimal algorithms for scheduling divisible workloads on heterogeneous systems. In: Proceedings of the 12th IEEE heterogeneous computing workshop
Dawson J, Strazdins P (1999) Optimizing user-level communication patterns on the Fujitsu AP3000. In: Proceedings of the 1st IEEE international workshop on cluster computing, pp 105–111
Bal HE, Plaat A, Bakker MG, Dozy P, Hofman RFH (1998) Optimizing parallel applications for wide-area clusters. In: Proceedings of the 12th international parallel processing symposium IPPS’98, pp 784–790
Faerman M, Birnbaum A, Casanova H, Berman F (2002) Resource allocation for steerable parallel parameter searches. In: Proceedings of GRID’02
Foster I, Kessclman C (1999) The grid: blueprint for a new computing infrastructure. Morgan Kaufmann, San Mateo. ISBN 1-55860-475-8
Frey J, Tannenbaum T, Livny M, Foster I, Tuccke S (2002) Condor-G: a computation management agent for multi-institutional grids. J Clust Comput 5:237–246
Guo M, Nakata I (2001) A framework for efficient data redistribution on distributed memory multicomputers. J Supercomput 20(3):243–265
Huang J-W, Chu C-P (2006) An efficient communication scheduling method for the processor mapping technique applied data redistribution. J Supercomput 37(3):297–318
Hsu C-H, Lin G-H, Li K-C, Yang C-T (2005) Localization techniques for cluster-based data grid. In: Proceedings of the 6th ICA3PP, Melbourne, Australia
Hsu C-H, Lo T-T, Yu K-M (2005) Localized communications of data parallel programs on multi-cluster grid systems. In: European grid conference. Lecture notes in computer science, vol 3470. Springer, Berlin, pp 900–910
Isaila F, Tichy WF (2002) Mapping functions and data redistribution for parallel files, In: Proceedings of IPDPS 2002 workshop on parallel and distributed scientific and engineering computing with applications, Fort Lauderdale, April
Javadi B, Akbari MK, Abawajy JH (2005) Performance analysis of heterogeneous multi-cluster systems. In: Proceedings of ICPP
Javadi B, Abawajy JH, Akbari MK (2005) Performance analysis of interconnection networks for multi-cluster systems. In: Proceedings of the 6th ICCS. Lecture notes in computer science, vol 3516. Springer, Berlin, pp 205–212
Knoop J, Mehofer E (2002) Distribution assignment placement: effective optimization of redistribution costs. IEEE Trans Parallel Distrib Syst 13(6):628–647
Kalns ET, Ni LM (1995) Processor mapping techniques toward efficient data redistribution. IEEE Trans Parallel Distrib Syst 6(12):1234–1247
Lee JS (2005) Data distribution management modeling and implementation on computational grid. In: Proceedings of the 4th GCC, Beijing, China
Lee S, Yook H-G, Koo M-S, Park M-S (2001) Processor reordering algorithms toward efficient GEN_BLOCK redistribution. In: Proceedings of the ACM 2001 symposium on applied computing
Lim YW, Bhat PB, Parsanna VK (1999) Efficient algorithm for block-cyclic redistribution of arrays. Algorithmica 24(3–4):298–330
Plaat A, Bal HE, Hofman RFH (1999) Sensitivity of parallel applications to large differences in bandwidth and latency in two-layer interconnects. In: Proceedings of the 5th IEEE high performance computer architecture HPCA’99, pp 244–253
Qin X, Jiang H (2001) Dynamic, reliability-driven scheduling of parallel real-time jobs in heterogeneous systems. In: Proceedings of the 30th ICPP, Valencia, Spain
Ranaweera S, Agrawal DP (2001) Scheduling of periodic time critical applications for pipelined execution on heterogeneous systems. In: Proceedings of the 30th ICPP, Valencia, Spain
Spooner DP, Jarvis SA, Caoy J, Saini S, Nudd GR (2003) Local grid scheduling techniques using performance prediction. IEE Proc Comput Digit Tech 150(2):87–96
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Hsu, CH., Chen, SC. Efficient selection strategies towards processor reordering techniques for improving data locality in heterogeneous clusters. J Supercomput 60, 284–300 (2012). https://doi.org/10.1007/s11227-010-0463-6
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11227-010-0463-6