THE “CPU+GPU+MIC” HYBRID SUPERCOMPUTER π
- Details
- Hits: 3166
Volume 1 (1), July 2018, Pages 126-128
James Lin1, Minhua Wen1
1Center for HPC, Shanghai Jiao Tong University, Shanghai, China, This email address is being protected from spambots. You need JavaScript enabled to view it., This email address is being protected from spambots. You need JavaScript enabled to view it.
Abstract
The supercomputer π at Center for HPC (CHPC) of Shanghai Jiao Tong University (SJTU) was the fastest supercomputer among all the universities in China during 2013 to 2015. With a highly hybrid architecture, it is the first “CPU+GPU+MIC” supercomputer in China. It was also the largest Kepler GPU cluster in China when deployed in 2013. In addition, it firstly utilized the “FDR+DDN” combination in China. In this paper, we will introduce the design of supercomputer π.
Keywords:
Supercomputer π, CPU, GPU, MIC, Hybrid Supercomputer
DOI: https://doi.org/10.32010/26166127.2018.1.1.126.128
Reference
[1]. Schulte, M. J. (2015) Achieving exascale capabilities through heterogeneous computing. IEEE Micro 35(4), 26-36.
[2]. Mittal, S., Jeffrey S. V. (2015) A survey of CPU-GPU heterogeneous computing techniques. ACM Computing Surveys (CSUR) 47(4), 69.
[3]. Jin, G., Lin, J., Endo, T. (2014) Efficient utilization of memory hierarchy to enable the computation on bigger domains for stencil computation in CPU-GPU based systems. Proceedings of the International Conference on High Performance Computing and Applications.
[4]. Sîrbu, A., Babaoglu, O. (2016) Power consumption modeling and prediction in a hybrid CPU-GPU-MIC supercomputer. European Conference on Parallel Processing.
[5]. Bientinesi, P. (2015) Parallel computing on graphics processing units and heterogeneous platforms. Concurrency and Computation: Practice and Experience 27(6), 1525-1527.
[6]. Sawadsitang, S. (2015) Understanding Performance Portability of OpenACC for Supercomputers. IEEE International Parallel and Distributed Processing Symposium Workshop.
[7]. Fu, H. (2017) Solving global shallow water equations on heterogeneous supercomputers. PloS one 12(3).
