StreamMAP: Automatic Task Assignment System on GPU Cluster

Chong Chong Sheng; Wei Song Hu; Xin Ming Hu; Bai Feng Wu

doi:10.4028/www.scientific.net/AMR.926-930.2414

Paper Titles

Research on Reader Service Mode in the Library Based on Digital Networked Environment
p.2398

Research on the Agent Oriented Software Design Engineering
p.2402

Research on the Development and Applications of Computer Science and Technology
p.2406

Research on the Development of Smart Library Based on Embedded RFID Middleware
p.2410

StreamMAP: Automatic Task Assignment System on GPU Cluster
p.2414

Structural Analysis and Preparation of the Netfilter
p.2418

Study On Development of Management Information System
p.2422

Computer Aided Sports Technology in Physical Training
p.2426

Application and Research Analysis of Computer in Modern Athletic Training
p.2430

HomeAdvanced Materials ResearchAdvanced Materials Research Vols. 926-930StreamMAP: Automatic Task Assignment System on GPU...

StreamMAP: Automatic Task Assignment System on GPU Cluster

Abstract:

GPU Clusters which use General-Purpose GPUs (GPGPUs) as accelerators are becoming more and more popular in high performance computing area. Currently the mainly used programming model for GPU cluster is hybrid MPI/CUDA. However, when using this model, programmers tend to need detailed knowledge of the hardware resources, which makes the program more complicated and less portable. In this paper, we present StreamMAP, an automatic task assignment system on GPU Clusters. The main contributions of StreamMAP are (1) It provides powerful yet concise language extension suitable to describe the computing resource demands of cluster tasks. (2) It maintains resource information and implements automatic task assignment for GPU Cluster. Experiments show that StreamMAP provides programmability, portability and performance gains.

You might also be interested in these eBooks

Progress in Applied Sciences, Engineering and Technology

View Preview

Info:

Periodical:

Advanced Materials Research (Volumes 926-930)

Pages:

2414-2417

DOI:

https://doi.org/10.4028/www.scientific.net/AMR.926-930.2414

Citation:

Cite this paper

Online since:

May 2014

Authors:

Chong Chong Sheng, Wei Song Hu, Xin Ming Hu, Bai Feng Wu

Keywords:

GPU Cluster, Hybrid MPI/CUDA, Portability, Task Assignment

Export:

RIS, BibTeX

Price:

Permissions CCC:

Request Permissions

Permissions PLS:

Request Permissions

Сopyright:

Citation:

References

[1] Information on http: /computing. ornl. gov/HMC/documents/HMCworkshopHwu. pdf.

Google Scholar

[2] J.C. Phillips, J.E. Stone and K. Schulten: Adapting a message-driven parallel application to gpu-accelerated clusters. In Proceedings of the 2008 ACM/IEEE conference on Supercomputing, SC'08, pages 8: 1-8: 9, Piscataway, NJ, USA, 2008. IEEE Press.

DOI: 10.1109/sc.2008.5214716

Google Scholar

[3] V.V. Kindratenko, J.J. Enos, G. Shi, et al: GPU clusters for high-performance computing. 2009 IEEE International Conference on Cluster Computing and Workshops. 2009, pp.1-8.

DOI: 10.1109/clustr.2009.5289128

Google Scholar

[4] S. Chakrabarti, J. Demmel, and K. Yelick: Modeling the benefits of mixed data and task parallelsm. In SPAA'95, pp.74-83.

DOI: 10.1145/215399.215423

Google Scholar

[5] T. Okuyama, F. I., and K. H: A task parallel algorithm for computing the costs of all-pairs shortest paths on the cuda-compatible gpu. In ISPA'08, page 284-291, (2008).

DOI: 10.1109/ispa.2008.40

Google Scholar

[6] Information on http: /www. nvidia. com/object/nvidia-kepler. html.

Google Scholar