Number of times this page has been accessed since 13 Jan. 2006:
Japanese version is here
1. Ph.D (in Computer Science) at the University of Tokyo since 29th March 2001.
2. Full Professor, High Performance Computing Division, Information Technology Center, Nagoya University, since 1st April, 2016. (Full Professor with
Tenure, since 1st April, 2022)
3. Full Professor, Department of Computer and Software Systems, Graduate School of Informatics, Nagoya University, since 1st April, 2017.
4. Full Professor, Department of Computer Science, School of Informatics, Nagoya University, since 1st April, 2017.
To whom wants to do a collaborative
research with our laboratory
In our laboratory, all topics for high performance computing are researching to be world-wide technology.
In particular, we focus on technologies for usage of advanced supercomputers, such as massively parallel algorithms, high performance implementations, and useful tools and software.
For Students:
In this laboratory, we conduct comprehensive research in high-performance computing, striving to achieve world-class results. Our focus is particularly on technological advancements related to cutting-edge supercomputers. We engage in research and development in areas such as massively parallel computing, large-scale machine learning (AI), quantum
computing technologies, GPU computing, high-performance implementation
techniques (GPU/multi-core optimization), and the development of practical
software tools.
*However, our research is not limited to these areas. If you have a specific topic you are interested in pursuing, please consult
with us in advance.
*Policy from 2024
In our laboratory, priority will be given to students who are motivated
to leverage the latest advancements in supercomputers and high-performance computing
(HPC) for research that fully utilizes artificial intelligence (AI) and focuses on enhancing the speed and performance (high quality, high accuracy) of real-world applications. Therefore, for master's students and above, it will be extremely challenging
to progress in research and complete a master's thesis unless they have
acquired foundational knowledge in supercomputer utilization, parallelization, and GPU computing
during their undergraduate studies. Please be sure to verify this requirement in advance.
** Graduate students:
(A Master Course)
Take an examination of Department of Computing and Sotware Systems, Graduate School of Informatics, Nagoya University. Then select Katagiri-Ogino laboratory as your request
for Master�fs research. In financial year of 2016, we accept 3 Master students.
(A Ph.D Candidate)
Take an examination of Department of Computing and Sotware Systems, Graduate School of Informatics, Nagoya University. It is strongly required for discussion with Prof.
Katagiri for your research in advance. Hence please contact to Prof. Katagiri
in advance.
(Postdocs)
Please contact to Prof. Katagiri.
*For Industries:
We accept collaborative researches between industries by applying framework
of collaborative research in Nagoya University to utilize supercomputers in Information Technology Center, Nagoya University.
The accepted researches should be limited due to human resources. If you
are interested in. please contact to Prof. Katagiri.
Attention: We are sorry in advance for
foreign industries in case that: Japanese low may affect or be restricted for
usage of supercomputers in our center to establish the collaboration.
What's new
"The Art of High Performance Computing for Computational Science, Vol. 1
:Techniques of Speedup and Parallelization for General Purposes" (Editor: Masaaki Geshi
"High-Performance Computing Basics", Takahiro Katagiri,
Pages 1-25
"Basics of MPI Programming",Takahiro Katagiri, Pages
27-44
"Basics of OpenMP Programming", Takahiro Katagiri,
Pages 45-59
"Hybrid Parallelization Techniques", Takahiro Katagiri,
Pages 61-68
"Application of Techniques for High-Performance Computing",
Takahiro Katagiri, Pages 69-81
DOI: https://doi.org/10.1007/978-981-13-6194-4
Springer, Singapore, Print ISBN 978-981-13-6193-7, Online ISBN 978-981-13-6194-4
Software Automatic Tuning, Naono, K.; Teranishi, K.; Cavazos, J.; Suda,
R. (Eds.),Springer, (September 29, 2010) was published.
pXabclib version Trial 0.10 was released. (17th, December, 2014)
ppOpen-HPC Home Page was renewal. (30th, June, 2012)
ppOpen-AT was released. (30th, June, 2012)
Japan Society of Promotion Science (JSPS) Bilateral Joint Research Projects
(Open Partnership, Japan-Taiwan), �gDeepening Performance Models for Automatic
Tuning with International Collaboration�h (Host place: Information Technology
Center, Nagoya University, FY 2016 - FY 2017)
�@First International Workshop on Deepening Performance Models for Automatic Tuning (DPMAT),
September 7th, 2016@Room IB-014, IB Building (Integrated Building), Higashiyama
Campus, Nagoya University
Please access my ResearchGate account if you want to know my recent publications.
Thesis
(1)Sanya Tangpongprasit, Takahiro Katagiri, Kenji Kise, Hiroki Honda, and Toshitsugu Yuba, Parallel Computing, Vol.31,
pp.529-543 (2005):
"A Time-To-Live Based Reservation
Algorithm of Fully Decentralized Resource Discovery in Grid Computing"
(2)Takahiro Katagiri, Kenji Kise, Hiroki Honda, and Toshitsugu Yuba, Parallel Computing, Vol.32, Issue 1, pp.92-112 (January 2006):
"ABCLibScript: A Directive to Support Specification of An Auto-tuning Facility
for Numerical Software"
(3)Takahiro Katagiri, Kenji Kise, Hiroki Honda, and Toshitsugu Yuba, Parallel Computing, Vol.32,
Issue 3, pp.231-250 (March 2006):
"ABCLib_DRSSED: A Parallel Eigensolver with an Auto-tuning Facility"
(1)Satoshi Ohshima, Kenji Kise, Takahiro Katagiri, and Toshitugu Yuba,
Selected Paper of VECPAR'2006, Springer LNCS 4395, pp.305--318, (2007):
"Parallel Processing of Matrix Multiplication
in a CPU and GPU Heterogeneous Environment"
(2)Takahiro Katagiri, Christof Voemel, and James Demmel, Selected Paper
of Workshop On State-of-the-art In Scientifc And Parallel Computing (PARA'06),
Springer LNCS 4699, pp.938--948, (2007):
"Automatic Performance Tuning for the
Multi-section with Multiple Eigenvalues Method for the Symmetric Eigenproblem"
(3)Teruo Tanaka, Takahiro Katagiri, and Toshitsugu Yuba, Selected Paper
of Workshop On State-of-the-art In Scientifc And Parallel Computing (PARA'06),
Springer LNCS 4699, pp.986--995, (2007):
"d-Spline Based Incremental Parameter
Estimation in Automatic Performance Tuning"
(4) Takahiro Katagiri, and Shoji Itoh, 9th International Meeting on High
Performance Computing for Computational Science (VECPAR'2010), Proceedings
of VECPAR'2010, 13 pages., Berkeley, California, USA, June 22-25 (2010):
"A Massively Parallel Dense Symmetric Eigensolver with Communication Splitting
Multicasting Algorithm"
(5) Takahiro Katagiri, Fifth International Workshop on Automatic Performance Tuning (iWAPT2010), Proceedings of VECPAR'2010, 2pages., Berkeley, California, USA, June 22 (2010):
"Challenges of Run-time Auto-tuning for Sparse Iterative Solvers" (A Position Paper)
Technical Reports
(1) Takahiro Katagiri, Jun'ichi Iwata, Kazuyuki Uchida: Technical Reports,
�@�@�@�@"A Communication Avoiding and Reducing Algorithm for Symmetric Eigenproblem
for Very Small Matrices"
(Research Gate Page
Selected Oral Presentations
(1)Takahiro Katagiri, 13th SIAM Conference on Parallel Processing for Scientific
Computing (PP08), Mini Symposium, MS1: Auto-tuning on Numerical Libraries
and Advanced Computer Systems: Part I of II, Atlanta, Georgia, USA, Wednesday,
March 12th, 2008
"Towards
General Auto-tuning Description Language on Advanced Computing Systems"
(PPT file)
(2) Takahiro Katagiri, Takao Sakurai, Hisayasu Kuroda, Ken Naono and Kengo
Nakajima, International Workshop on Peta-Scale Computing Programming Environment,
Languages and Tools (WPSE 2010), Kyoto, February 18, 2010
"Xabclib:
An Iterative Solver with a General Auto-tuning Interface "OpenATLib"
(3) Takao Sakurai, Ken Naono, Hisayasu Kuroda, Takahiro Katagiri, and Kengo
Nakajima, 14th SIAM Conference on Parallel Processing for Scientific Computing
(PP2010), Joint JSIAM -- SIAM Minisymposium: The State-of-the-art of Auto-tuning
Technologies: Adaptation to Advanced Computer Environment and Numerical
Libraries - Part I of II, MS6, Grand Hyatt Seattle, Seattle, Washington,
USA, Wednesday, February 24, 2010
"OpenATLib:
A General Auto-tuning Interface for Numerical Solvers "
(4) Takahiro Katagiri, Hisayasu Kuroda and Kengo Nakajima, 14th SIAM Conference
on Parallel Processing for Scientific Computing (PP2010), Joint JSIAM --
SIAM Minisymposium: Parallel Programming Models and Algorithms for Multicore
Clusters and GPGPUs - Part III of III, MS62, Grand Hyatt Seattle, Seattle,
Washington, USA, Wednesday, February 26, 2010
"A Branchless
Segmented Scan Method for Sparse Matrix-vector Multiplications "
(5) Satoshi Ohshima, Takahiro Katagiri, Kengo Nakajima, Hisayasu Kuroda
and Shoji Itoh, International Conference on Suerconputing (ICS'2010), Epocal
Tsukuba, Tsukuba, Japan, 1st June-4th June, 2010
"First Implementation
of Segmented Scan Method on GPU" (A Poster Presentation)
(6)Takahiro Katagiri, Takao Sakurai, Hisayasu Kuroda, Ken Naono, and Kengo
Nakajima, 2nd International Workshops on Advances in
Computational Mechanics (IWACOM-II) (March 29-31, 2010, Yokohama, JAPAN),
OW-1 Innovative Algorithms for Multicore and Hierarchical Computational
Environments (Organizers: H. Okuda, K. Nakajima and T. Aoki), March, 31 (Wed)
Room A, [OW-1: Session-5] 14:20-14:50
"Development
of numerical computation policy facility and its effect on Xabclib"
(7)Takahiro Katagiri, Takao Sakurai, Satoshi Ohshima, Mitsuyoshi Igai,
Hisayasu Kuroda, Kengo Nakajima, Ken Naono, Ken Naono and Shoji Itoh, 2010 SIAM
Annual Meeting, (July 12-16, 2010, The David Lawrence Convention Center,
Pittsburgh, PA), MS105, Architecture-Aware Algorithms - Part I of II, 10:30 AM -
12:30 PM, Room: Spirit of Pittsburgh B - Level 3, (Organizer: Xiaoye S. Li
(Lawrence Berkeley National Laboratory))
"Numerical Policy
Function on Auto-tuned Sparse Iterative Solver Xabclib "
�i8�jTakahiro Katagiri, Junichi Iwata, and Kazuyuki Uchida, MS34: Auto-tuning Technologies for Extreme-Scale Solvers - Part I of III, Organizers: Takahiro Katagiri (The University of Tokyo, Japan), Osni A. Marques (Lawrence Berkeley National Laboratory, USA), Toshiyuki Imamura (RIKEN, Japan), and Leroy A. Drummond (Lawrence Berkeley National Laboratory, USA), SIAM Conference on Parallel Processing (PP14), Marriott Portrand, Downtown Waterfront, Portland, Oregon, USA, February 18-21, 2014
�@�@�@"Extreme-Scale Parallel Symmetric Eigensolver for Very Small-Size
Matrices Using A Communication-Avoiding for Pivot Vectors " (Oral
Presentation)
Online Presentation by SIAM
PPT by my Slide Share
Professor Takahiro Katagiri received his Associate Bachelor of Engineering (B.E.) degree from Toyota National College of Technology in 1994, his B.E. from Kyoto University in 1996, his M.Sc. degree from the University of Tokyo in 1998, and his Ph.D (in Computer Science) degree from the University of Tokyo in 2001.
His main research interests focuses on parallel matrix calculation, High
Performance Computing, and software auto-tuning. He is presently professor,
High Performance Computing Division, Information Technology Center, Nagoya
University.
After receiving Ph.D degree, he was a project lead of Japan Science and Technology in 2001, and he moved to Graduate School of Information Systems, The University of Electro-Communications in 2002 as a research associate.
He was also a visiting scholar at University of California at Berkeley (UCB) in Computer Science Division until January 2006. The supervised professors in UCB are Prof. Katherine Yelick and Prof. James Demmel, and the host research group is Berkeley Benchmarking and OPtimization (BeBOP) group.
In 2007, he moved to Supercomputing Research Division, Information Technology
Center, the University of Tokyo as an associate professor. In 2016, he
moved to current position.
He is a developer of Automatically Blocking-and-Communication adjustment Library (ABCLib), ABCLibScript which is a support language for automatic code generation for numerical libraries with auto-tuning facility. He is a proponent of Framework of Install-time, Before Execute-time, and Run-time optimization layers (FIBER) for the software auto-tuning framework.
In 2011, we received Commendation for Science and Technology by the Minister of Education, Culture, Sports, Science and Technology, Young Scientist Award for contribution of research on auto-tuning.
He is a member of ACM, IEEE, SIAM, IPSJ and JSIAM.
How to contact:
* Furo-cho, Chikusa-ku, Nagoya, 464-8601, Japan
Information Technology Center, Nagoya
University,
5F Room 505
High Performance Computing Division,
Information Technology Center,
Nagoya University
[Last updated] 25, September, 2024