| Takeshi Nanri | Last modified date：2022.05.13 |
Associate Professor /
Department of Advanced Information Technology, Faculty of Information Science and Electrical Engineering
Section of Advanced Computational Science
Research Institute for Information Technology
Section of Advanced Computational Science
Research Institute for Information Technology
E-Mail *Since the e-mail address is not displayed in Internet Explorer, please use another web browser:Google Chrome, safari.
Homepage
https://kyushu-u.pure.elsevier.com/en/persons/takeshi-nanri
Reseacher Profiling Tool Kyushu University Pure
http://aqua.cc.kyushu-u.ac.jp/~nanri/
Takeshi Nanri .
Fax
092-802-2642
Academic Degree
Ph.D
Country of degree conferring institution (Overseas)
No
Field of Specialization
High-performance computing
Total Priod of education and research career in the foreign country
00years00months
Outline Activities
Research:
My major research topic is to develop technologies for runtime systems of parallel programs.
Parallel computers have become popular platform from PC servers to large-scale supercomputers. To utilize the computational power of those platforms, users must prepare parallel programs. In an execution of a parallel program on a parallel computer, a runtime system interprets the operation written in the program into specific behavior of the computer. Therefore, technologies in runtime systems are the keys for efficient use of parallel computers. Especially, when the size of the computer becomes large, to achieve better performance, consideration of the information available at runtime, such as the allocation of processes, balance of load among processors or contentions of resources among jobs. In our research, we are developing technologies for dynamic optimization of runtime systems on large-scale parallel computers.
My major research topic is to develop technologies for runtime systems of parallel programs.
Parallel computers have become popular platform from PC servers to large-scale supercomputers. To utilize the computational power of those platforms, users must prepare parallel programs. In an execution of a parallel program on a parallel computer, a runtime system interprets the operation written in the program into specific behavior of the computer. Therefore, technologies in runtime systems are the keys for efficient use of parallel computers. Especially, when the size of the computer becomes large, to achieve better performance, consideration of the information available at runtime, such as the allocation of processes, balance of load among processors or contentions of resources among jobs. In our research, we are developing technologies for dynamic optimization of runtime systems on large-scale parallel computers.
Research
Research Interests
Membership in Academic Society
- Fundamental techniques to enable highly scalable parallel computations
keyword : scalability, parallel computation, high-performance computation
2011.09. - Technologies for dynamic optimization of communication libraries on large-scale parallel computers
keyword : Parallel Computing, Runtime Optimization
2005.04Study for speeding up MPI programs by adjusting the implementation of communication algorithms or the allocation of ranks according to the system performance of parallel machines, or the status of workloads on machines.. - Programming environment for hierarchical parallel environment
keyword : Hierarchical parallel computer, distributed shared memory, communication optimization
2003.04.
- Within a decade, the number of processing cores on supercomputers is predicted to be more than 100 million. This project researches technologies for memory saving and runtime optimizations to implement a scalable communication library that will be required on such large scale computers. In addition to that, the project also develops methods for building scalable applications by utilizing facilities of the communication library.
- PSI is one of the national projects on elemental technologies for peta-scale computing systems. The project works intensively on the topics of system interconnection networks: Optical switches, Intelligent interconnects and Performance prediction.
Papers
- IEEE
Educational
Unauthorized reprint of the contents of this database is prohibited.