Benchmarking: Performance on ALPS and Formosa Clusters

Authors: Chih-Wei Hsieh, Chau-Yi Chou, Sheng-HsiuKuo, Tsung-Che Tsai, I-Chen Wu

Abstract:

This paper presents the benchmarking results and performance evaluation of differentclustersbuilt atthe National Center for High-Performance Computingin Taiwan. Performance of processor, memory subsystem andinterconnect is a critical factor in the overall performance of high performance computing platforms. The evaluation compares different system architecture and software platforms. Most supercomputer used HPL to benchmark their system performance, in accordance with the requirement of the TOP500 List. In this paper we consider system memory access factors that affect benchmark performance, such as processor and memory performance.We hope these works will provide useful information for future development and construct cluster system.

Keywords: Performance Evaluation, Benchmarking and High-Performance Computing

Digital Object Identifier (DOI): doi.org/10.5281/zenodo.1056464

References:

[1] Chau-Yi Chou, Hsi-Ya Chang, Shuen-Tai Wang, Te-Min Chen, and Chang-Hsing Wu, "Benchmarking and Performance Analysis on a Blade Cluster,"The 3rd Workshop on Grid Technologies and Applications (WoGTA), 2006.
[2] Kuo-Chan Huang, Hsi-Ya Chang, Cherng-YeuShen,Chaur-Yi Cho andShou-Cheng Tcheng,"Benchmarking and performance evaluation of NCHC PC cluster,"The Fourth International Conference/Exhibition on HighPerformance Computing in Asia-Pacific Region (HPC Asia 2000 Conference), pp. 923-928, Beijing, China, May 14-17, 2000.
[3] Chin-Hung Li, Te-Ming Chen, Ying-Chuan Chen, and Shuen-Tai Wang, "Formosa3: A Cloud-Enabled HPC Cluster in NCHC," International Conference on Electrical, Computer, Electronics and Communication Engineering, 2011.
[4] P.J. Mucci and K.S. London, "Low Level Architectural Characterization Benchmarks for Parallel Computers," Technical Report UT-CS-98-394, Univ. of Tennessee, 1998.
[5] Douglas M. Pase, "Linpack HPL Performance on IBM eServer 326 and xSeries 336 Servers," IBM, 2005.
[6] T. Sterling, D. Becker, D. Savareseet. al., "BEOWULF: A ParallelWorkstation for Scientific Computation," Proc. Of International Conf: On Parallel Processing (ICPP), August 1995.
[7] HPL, Available at: http://www.netlib.org/benchmark/hpl/.
[8] TOP500 Lists, Available at : http://top500.org
[9] The Message Passing Interface (MPI) standard, Available at: http://www.mcs.anl.gov/research/projects/mpi/