November 19, 2014
University of Tsukuba
Top honors awarded for productivity in a high-performance supercomputer parallel programming language implementation and overall performance
RIKEN, the University of Tsukuba and Fujitsu today announced that they received top ranks in two of the four benchmarks at the 2014 HPC Challenge Class 1 Awards, which evaluate the overall performance of supercomputers, for the performance of the K computer. The first-place rankings were received in two benchmarks: (1) Global HPL, which measures the floating point rate of execution for solving a linear system of equations; and (2) EP STREAM (Triad) per system, which measures sustainable memory bandwidth and the corresponding computation rate for simple vector kernels. This marks the fourth consecutive year, from 2011 to 2014, that the K computer has ranked first in the HPC Challenge Class 1 Awards.
In addition, XcalableMP and XcalableACC, which are high-performance supercomputer parallel programming languages developed by RIKEN and the University of Tsukuba, received award in the HPC Challenge(1) Class 2 Best Performance Award, which recognize the overall performance of a programming language. This is the second year running that XcalableMP has won this award.
The awards were announced on November 18, 2014, in New Orleans, Louisiana, at SC14, the International Conference for High Performance Computing, Networking, Storage and Analysis.
The HPC Challenge benchmarks are benchmark programs designed to evaluate the overall performance of supercomputers in terms of processing performance based on 28 frequently used computational patterns in the field of scientific computation. At the contest using the HPC Challenge benchmarks, which takes place once a year, there are two classes of awards: Class 1, which measures benchmark performance values, and Class 2, which measures the productivity of programming language implementations.
Among the HPC Challenge benchmarks, particularly important are (1) Global HPL; (2) EP STREAM (Triad) per system; (3) Global FFT; and (4) Global RandomAccess, and top rankings are awarded in each category in the HPC Challenge Class 1 Awards. RIKEN, the University of Tsukuba, and Fujitsu registered performance for each of these benchmarks in the HPC Challenge Class 1 Awards. The University of Tsukuba made especially important contributions to increasing the processing speed for Global FFT.
The HPC Challenge Class 2 Award is a contest for programming languages used in developing HPC applications. This award is designed to evaluate both programming language productivity and performance for four HPC Challenge benchmarks in the HPC Challenge Class 1 Awards.
The award-winning XcalableMP is a programming language that was jointly developed by the RIKEN Advanced Institute for Computational Science and the University of Tsukuba’s Center for Computational Sciences. XcalableACC, which simultaneously won the award, is an extension of the XcalableMP programming language that enables applications to be developed for supercomputers that are equipped with accelerators. In a performance evaluation of the HPC Challenge benchmarks of XcalableMP using the K computer, extraordinarily high productivity and performance was demonstrated. In addition, the performance of XcalableACC was evaluated using the University of Tsukuba’s HA-PACS) supercomputer equipped with a GPU accelerator, and it, too, demonstrated exceptionally high productivity and performance.
Programming languages that can be used to develop applications in a short period of time while maintaining performance and that run on large-scale computation environments, such as the K computer, make it possible to accelerate the pace of research. As a result, they are in high demand by researchers both inside and outside Japan. The awards reveal both the high productivity and high performance of XcalableMP and XcalableACC, in addition to demonstrating the substantial effectiveness of programming languages for developing HPC applications.
The top three rankings achieved on the four benchmarks for the HPC Challenge Class 1 Awards for 2014 are as follows:
|Global HPL||Performance (TFLOP/s)||System||Institutional Facility|
|1st place||9,796||K computer||RIKEN|
|2nd place||5,709||Mira||Argonne National Laboratory|
|EP STREAM(Triad) per system||Performance (TB/s)||System||Institutional Facility|
|1st place||3,857||K computer||RIKEN|
|2nd place||1,426||Mira||Argonne National Laboratory|
|Global FFT||Performance (TFLOP/s)||System||Institutional Facility|
|1st place||226||Mira||Argonne National Laboratory|
|2nd place||206||K computer||RIKEN|
|Global RandomAccess||Performance (GUPS)||System||Institutional Facility|
|1st place||2,021||Power 775||IBM|
|2nd place||472||K computer||RIKEN|
|3rd place||418||Mira||Argonne National Laboratory|
For more information:
- HPC Challenge Benchmark: http://icl.cs.utk.edu/hpcc/index.html
- RIKEN Advanced Institute for Computational Science: http://www.aics.riken.jp/en/
- Center for Computational Sciences, University of Tsukuba: https://www.ccs.tsukuba.ac.jp/eng/
- Fujitsu’s website on the K computer: http://www.fujitsu.com/global/about/businesspolicy/tech/k/
 HPC Challenge Awards
The HPC Challenge benchmarks are benchmark programs designed to evaluate the overall performance of supercomputers in terms of processing performance in 28 tests derived from frequently used computational patterns in the field of scientific computation. The HPC Challenge Awards are evaluated based on the HPC Challenge benchmarks. There are two classes of awards: Class 1, which measures benchmark performance values, and Class 2, which measures the productivity of programming language implementations. The Class 1 awards consist of the following four benchmarks, each of which evaluates the performance of key system components (CPU computational performance, memory access performance, network transmission performance).
– Global HPL: operating speed in solving large-scale simultaneous linear equations
– EP STREAM (Triad) per system: memory access speed under multiple loads
– Global FFT: total performance of Fast Fourier Transform
– Global RandomAccess: random memory access performance in parallel processing
HPC Challenge Class 2 Awards are given to programming languages used for developing HPC applications. Three benchmarks are selected from the four used in the Class 1 category, and both the productivity of the programming language and benchmark performance are evaluated for the implementation of these benchmarks. In addition, participants can choose to include up to two additional benchmarks besides the HPC Challenge benchmarks for consideration, and the award is determined based on the total score for the implementations including the additional benchmarks of Best Performance and Most Elegant Language or Most Elegant Implementation.
 K computer
The K computer, which was jointly developed by RIKEN and Fujitsu, is part of the High-Performance Computing Infrastructure (HPCI) initiative led by Japan’s Ministry of Education, Culture, Sports, Science and Technology (MEXT). The K computer’s availability for shared use began in September 2012. The “K” in K computer comes from the Japanese Kanji character “Kei” which means ten peta or 10 to the 16th power. In its original sense, “Kei” expresses a large gateway, and it is hoped that the system will be a new gateway to computational science.
A programming language that facilitates the development of parallel applications that run on supercomputers and other large-scale computational environments. With XcalableMP, it is possible to develop parallel applications that can achieve the same computational speeds as existing methods while using comparatively shorter code.
A programming language that is an extension of XcalableMP and OpenACC, the standard language for programming for accelerators. In addition to the conventional XcalableMP structure, it has a structure in which communication is conducted between accelerators, making it easy to develop applications that run on parallel computers that are equipped with accelerators. Development of the XcalableACC was supported by the JST/CREST program entitled “Research and Development on Unified Environment of Accelerated Computing and Interconnection for Post-Petascale Era” in the research area of “Development of System Software Technologies for Post-Peta Scale High Performance Computing.”
A supercomputer developed by the University of Tsukuba that is equipped with a GPU (Graphics Processing Unit) accelerator, which is the most representative type of accelerator. In the Green500 contest, which ranks supercomputers by how efficiently they can compute using the least amount of electrical power, HA-PACS ranked third in the world (as of June 2014) for its Tightly Coupled Accelerators (TCA).
RIKEN Advanced Institute for Computational Science
Office for Research Communications
E-mail: aics-koho [at] riken.jp
University of Tsukuba
Bureau of Public Relations, Center for Computational Sciences
E-mail: pr [at] ccs.tsukuba.ac.jp
Public and Investor Relations Division
RIKEN is Japan’s flagship research institute devoted to basic and applied research. Over 2500 papers by RIKEN researchers are published every year in reputable scientific and technical journals, covering topics ranging across a broad spectrum of disciplines including physics, chemistry, biology, medical science and engineering. RIKEN’s advanced research environment and strong emphasis on interdisciplinary collaboration has earned itself an unparalleled reputation for scientific excellence in Japan and around the world. For more information, please see: http://www.riken.jp/en
About University of Tsukuba
The University of Tsukuba aims to establish free exchange and close relationship in both basic and applied sciences with educational and research organizations and academic communities in Japan and overseas. The university makes a contribution to the world through its educational system that seeks to make the most of students’ creativity and individuality http://www.tsukuba.ac.jp/english/ https://www.ccs.tsukuba.ac.jp/eng/
Fujitsu is the leading Japanese information and communication technology (ICT) company offering a full range of technology products, solutions and services. Approximately 162,000 Fujitsu people support customers in more than 100 countries. We use our experience and the power of ICT to shape the future of society with our customers. Fujitsu Limited (TSE: 6702) reported consolidated revenues of 4.8 trillion yen (US$46 billion) for the fiscal year ended March 31, 2014. For more information, please see: http://www.fujitsu.com
All company or product names mentioned herein are trademarks or registered trademarks of their respective owners. Information provided in this press release is accurate at time of publication and is subject to change without advance notice.