Abstract
This chapter gives an overview of techniques for high-performance computing with actual examples of adaptations. In addition, several techniques of speedups for communications are also shown.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Notes
- 1.
There are many communication modes for MPI. For the default communication mode, the sending process is finished before the corresponding received calls if there is enough buffer of communications. Hence if the buffer of communications is small, sending does not follow this behavior. Thus, the sending should be held until calling the corresponding receiver.
- 2.
The Fujitsu PRIMEHPC FX100 system, which is installed in Information Technology Center, Nagoya University, has been supporting the overlapping between communications and computations by hardware, named assistant core , for dedicated hardware (CPU) for communications.
- 3.
In this case, users should write a directive or describe a compiler option to know nondependency for the argument values to compilers. The readers can survey these directives and options in their compilers before using it.
References
T. Furumura, L. Chen, Parallel Comput. 31, 149 (2005)
D.A. Patterson, J.L. Hennessy, in Computer Organization and Design MIPS Edition, Fifth Edition: The Hardware/Software Interface. The Morgan Kaufmann Series in Computer Architecture and Design (2013)
T. Katagiri, K. Kise, H. Honda, T. Yuba, in The Fifth International Symposium on High Performance Computing (ISHPC-V). Springer Lecture Notes in Computer Science, vol. 2858 (2003), p. 146
T. Tanaka, R. Otsuka, A. Fujii, T. Katagiri, T. Imamura, Scientific Program. 22, 299 (2014). IOS Press
R. Murata, J. Irie, A. Fujii, T. Tanaka, T. Katagiri, in Proceedings of Embedded Multicore/Many-core Systems-on-Chip (MCSoC) (2015), p. 203
T. Katagiri, K. Kise, H. Honda, T. Yuba, Parallel Comput. 32, 92 (2010)
T. Katagiri, S. Ohshima, M. Matsumoto, in Proceedings of IEEE MCSoC-2014, Special Session: Auto-Tuning for Multicore and GPU (ATMG-14) (2014), p. 91
T. Katagiri, S. Ohshima, M. Matsumoto, in Proceedings of IEEE IPDPSW 2015 (2015), p. 1221
T. Katagiri, S. Ohshima, M. Matsumoto, in Proceedings of IEEE IPDPSW 2016 (2016), p. 1488
T. Katagiri, S. Ohshima, M. Matsumoto, in Proceedings of IEEE IPDPSW 2017 (2017), p. 1399
ppOpen-HPC Project Home Page, http://ppopenhpc.cc.u-tokyo.ac.jp/ppopenhpc/
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2019 Springer Nature Singapore Pte Ltd.
About this chapter
Cite this chapter
Katagiri, T. (2019). Application of Techniques for High-Performance Computing. In: Geshi, M. (eds) The Art of High Performance Computing for Computational Science, Vol. 1. Springer, Singapore. https://doi.org/10.1007/978-981-13-6194-4_5
Download citation
DOI: https://doi.org/10.1007/978-981-13-6194-4_5
Published:
Publisher Name: Springer, Singapore
Print ISBN: 978-981-13-6193-7
Online ISBN: 978-981-13-6194-4
eBook Packages: Computer ScienceComputer Science (R0)