Kyushu University Academic Staff Educational and Research Activities Database
List of Presentations
Takeshi Nanri Last modified date:2018.04.24

Associate Professor / Advanced Computing Infrastructure Section / Research Institute for Information Technology


Presentations
1. Takeshi Nanri, Proposal of Interface for Runtime Memory Manipulation of Applications via PGAS-based Communication Library, Workshop on PGAS programming models: Experiences and Implementations (PGAS-EI), 2018.01.
2. Tetsuya Nakatoh, Sachio Hirokawa, Toshiro Minami, Takeshi Nanri, Miho Funamori, Attribute-based quality classification of academic papers, 2017.11, Investigating the relevant literature is very important for research activities. However, it is difficult to select the most appropriate and important academic papers from the enormous number of papers published annually. Researchers search paper databases by combining keywords, and then select papers to read using some evaluation measure—often, citation count. However, the citation count of recently published papers tends to be very small because citation count measures accumulated importance. This paper focuses on the possibility of classifying high-quality papers superficially using attributes such as publication year, publisher, and words in the abstract. To examine this idea, we construct classifiers by applying machine-learning algorithms and evaluate these classifiers using cross-validation. The results show that our approach effectively finds high-quality papers..
3. Shinji Sumimoto, Yuichiro Ajima, Takafumi Nose, Kazushige Saga, Naoyuki Shida, Takeshi Nanri, Parallel Application Experiences Using Advanced Communication Primitives, 25th Euromicro International Conference on Parallel, Distributed and network-based Processing, 2017.03.
4. Tetsuya Nakatoh, Kenta Nagatani, Toshiro Minami, Sachio Hirokawa, Takeshi Nanri, Miho Funamori, Analysis of the quality of academic papers by the words in abstracts, Thematic track on Human Interface and the Management of Information, held as part of the 19th International Conference on Human–Computer Interaction, HCI International 2017, 2017.01, The investigation of related research is very important for research activities. However, it is not easy to choose an appropriate and important academic paper from among the huge number of possible papers. The researcher searches by combining keywords and then selects an paper to be checked because it uses an index that can be evaluated. The citation count is commonly used as this index, but information about recently published papers cannot be obtained. This research attempted to identify good papers using only the words included in the abstract. We constructed a classifier by machine learning and evaluated it using cross validation. As a result, it was found that a certain degree of discrimination is possible..
5. Kazuichi Oe, Takeshi Nanri, Koji Okamura, Feasibility study for building hybrid storage system consisting of non-volatile DIMM and SSD, 4th International Symposium on Computing and Networking, CANDAR 2016, 2017.01, Various vendors develop a byte accessible Nonvolatile Dual-Inline Memory Module (NVDIMM). The performance of the NVDIMM drastically surpasses that of the Solid State Drive (SSD), which is connected by PCI express. However, the cost of the NVDIMM is much higher than that of the SSD. Therefore, a hybrid storage system between the NVDIMM and SSD is an effective technique for improving cost-performance. If a system uses the NVDIMM less while maintaining performance, its cost-performance should be improved. Our previous work involves on-the-fly automated storage tiering (OTF-AST). OTF-AST is a hybrid storage system consisting of an SSD and HDD. It aims to reduce the average response time of IO accesses by migrating only the IO concentration area to the SSD when IO concentration happens. Therefore, we construct OTF-AST with both the DIMM and SSD and evaluate it in order to understand how to build a cost-effective hybrid storage system with these devices. We use a DIMM instead of a byte accessible NVDIMM, which is difficult to obtain. As a result, we found that the original OTF-AST is suitable for a hybrid storage system consisting of the DIMM and SSD. Moreover, we can improve the performance of OTF-AST if replace its migration algorithm with a more positive migration algorithm. This is because the IO access response time barely increases when the data migration between the DIMM and SSD is done. We will build a more positive migration algorithm in the near future..
6. Shinji Sumimoto, Yuichiro Ajima, Kazushige Saga, Takafumi Nose, Naoyuki Shida, Takeshi Nanri, The design of advanced communication to reduce memory usage for exa-scale systems, 12th International Conference on High Performance Computing for Computational Science, VECPAR 2016, 2017.01, Current MPI (Message Passing Interface) communication libraries require larger memories in proportion of the number of processes, and can not be used for exa-scale systems. This paper proposes a global memory based communication design to reduce memory usage for exa-scale communication. To realize exa-scale communication, we propose true global memory based communication primitives called Advanced Communication Primitives (ACPs). ACPs provide global address, which is able to use remote atomic memory operations on the global memory, RDMA (Remote Direct Memory Access) based remote memory copy operation, global heap allocator and global data libraries. ACPs are different from the other communication libraries because ACPs are global memory based so that house keeping memories can be distributed to other processes and programmers explicitly consider memory usage by using ACPs. The preliminary result of memory usage by ACPs is 70 MB on one million processes..
7. Hiroaki Honda, Yoshiyuki Morie, Takeshi Nanri, Development of A Memory Efficient Communication Method for Connecting MPI Programs by using ACP Library, 5th JSST Annual Conference International Conference on Simulation Technology, 2016.10.
8. Ryutaro Susukita, Yoshiyuki Morie, Takeshi Nanri, Efficient communications of particle data in particle-based simulations, 5th JSST Annual Conference International Conference on Simulation Technology, 2016.10.
9. Keiichiro Fukazawa, Takayuki Umeda, Takeshi Nanri, Performance Evaluation of MHD Simulation Code with X86 CPUs and Manycore Systems, 5th JSST Annual Conference International Conference on Simulation Technology, 2016.10.
10. Takeshi Nanri, Keiichiro Fukazawa, Effect of Overlapping Halo Exchange with One-Sided Communication, 5th JSST Annual Conference International Conference on Simulation Technology, 2016.10.
11. Seiji FUJINO, Takeshi Nanri, Improvement of Eisenstat-SSOR preconditioning using tolerance value, 5th IMA Conference on Numerical Linear Algebra and Optimization, 2016.09.
12. Keiichiro Fukazawa, Toshiya Takami, Takeshi Soga, Yoshiyuki Morie, Takeshi Nanri, Effective Calculation with Halo communication using Halo Functions, 23rd European MPI Users' Group Meeting, 2016.09.
13. Shinji Sumimoto, Yuichiro Ajima, Kazushige Saga, Takafumi Nose, Naoyuki Shida, Takeshi Nanri, The Design of Advanced Communication to Reduce Memory Usage for Exa-scale Systems, 12th International Meeting On High Performance Computing for Computational Science, 2016.09.
14. Keiichiro Fukazawa, Yoshiyuki Morie, Toshiya Takami, Takeshi Nanri, Takeshi Soga, Effective calculation with halo communication using halo functions, 23rd European MPI Users' Group Meeting, EuroMPI 2016, 2016.09, The issue of halo communication is the decrease of parallel scalability. To overcome the issues, we have introduced "Halo thread" to our simulation code. However, we have not solved the issue basically in the strong scaling. In this study, we have developed the Halo functions which perform the halo communication effectively. Then we can perform the calculation and communication in a pipeline and obtained good performance..
15. Takeshi Nanri, Runtime Algorithm Selection of Collective Communication with RMA-based Monitoring Mechanism, 4th Annual MVAPICH Users Group Meeting, 2016.08.
16. Yoshiyuki Morie, Hiroaki Honda, Takeshi Nanri, Taizo Kobayashi, Hidetomo Shibamura, Ryutaro Susukita, Yuichiro Ajima, Memory Efficient One-Sided Communucation Library "aCP" in Globary Memory on Raspberry Pi 2, 36th IEEE International Conference on Distributed Computing Systems, ICDCS 2016, 2016.08, Previously, communications in parallel programs forHigh Performance Computing (HPC) and Distributed Computing(DC) are mostly written with two-sided communicationinterfaces that are based on a pair of operations, Send andReceive. Since such interface requires explicit synchronizationbetween both sides of the communication, techniquesfor communication optimization such as overlapping are notefficiently described in many cases. On the other hand, onesidedcommunication interface is becoming important as amethod to describe asynchronous communications to enablehighly overlapped communication with computation. As oneof such interface, in this demonstration, Advanced CommunicationPrimitives (ACP) is introduced. ACP is a portableinterface that supports UDP, IBverbs of InfiniBand and Tofulibrary of K Computer. In addition to that, it is designed tobe memory efficient. For example, with 10 thousand processes, the memory consumption of ACP over UDP is estimated to beless than 1MB. Since the number of computational elements isincreasing more rapidly than the amount of available memory, this memory efficiency is becoming one of the keys for parallelprograms in HPC and DC. To show this characteristics, we runACP library on Raspberry Pi 2, and examine its performanceand memory consumption..
17. Ryutaro Susukita, Yoshiyuki Morie, Takeshi Nanri, NSIM-ACE: An Interconnection Network Simulator for Evaluating Remote Direct Memory Access, International Conference on Simulation and Modeling Methodologies, Technologies and Applications, 2016.07.
18. Takeshi Nanri, Evaluation of On-Demand Message-Passing Module over RDMA Network, ACSI2016, 2016.01.
19. Kazuichi Oe, Takeshi Nanri, KOJI OKAMURA, Analysis of Storage Workloads of Input-Output Access Locality and Designing of Hybrid Storage System, 1st International Conference on Enterprise Architecture and Information Systems, 2016.01.
20. Ryutaro Susukita, Yoshiyuki Morie, Takeshi Nanri, Hidetomo Shibamura, Performance Evaluation of RDMA Communication Patterns by Means of Simulations, 2015 Joint International Mechanical, Electronic and Information Technology Conference, 2015.12.
21. Kazuichi Oe, Takeshi Nanri, KOJI OKAMURA, On-The-Fly Automated Storage Tiering with Caching and both Proactive and Observational Migration, Workshop on Computer Systems and Architectures (CSA'15), 2015.12.
22. Hiroaki Honda, Takeshi Nanri, Yoshiyuki Morie, Performance and memory usage evaluations for channel interface of Advanced Communication Primitives library, 1st Pan-American Congress on Computational Mechanics (PANACM 2015), 2015.04.
23. Takeshi Nanri, Channel Interface: a Primitive Model for Memory Efficient Communication, 23rd Euromicro International Conference on Parallel, Distributed and network-based Processing, 2015.03.
24. Takeshi Nanri, Design and Implementation of Channel Interface as a Memory Efficient Communication Model, Annual Meeting on Advanced Computing System and Infrastructure (ACSI) 2015, 2015.01.
25. Takeshi Nanri, Proposal of HINT Interface for Runtime Tuning of Communication Links, 22nd Euromicro International Conference on Parallel, Distributed and network-based Processing, 2014.02.
26. Tsuyoshi Okuma, Takeshi Nanri, Performance Study of Non-blocking Collective Communication Implementations Toward Adaptive Selection, Networking, Computing, Systems and Software, 2013.12.
27. Hironobu Sugiyama, Takeshi Nanri, Topology Aware Performance Prediction of Collective Communication Algorithms on Multi-Dimensional Mesh/Torus, Networking, Computing, Systems and Software, 2013.12.
28. Yoshiyuki Morie, Takeshi Nanri, A neighbor communication algorithm with making an effective use of NICs on multidimensional-mesh/torus, International Conference on Simulation Technology , 2013.09.
29. Takeshi Nanri, What Communication Library Can do with a Little Hint from Programmers?, MVAPICH User Group Meeting, 2013.08.
30. Takeshi Nanri, Hironobu Sugiyama, FUKAZAWA Keiichiro, A Cost-Efficient Approach for Automatic Algorithm Selection of Collective Communications, SIAM Conference on Computational Science and Engineering, 2013.03.
31. Takeshi Nanri, Introduction of ACE(Advanced Communication library for Exa) Project, International workshop on HPC, Krylov Subspace method and its application, 2013.01.
32. Yoshiyuki Morie, Takeshi Nanri, Task Allocation Method for Avoiding Contentions by the Information of Concurrent Communication, International workshop on HPC, Krylov Subspace method and its application, 2013.01.
33. Tsuyoshi Okuma, Takeshi Nanri, Evaluation of Implementation Methods for Non-Blocking Collective Communications in Overlapping Communication and Computation, International workshop on HPC, Krylov Subspace method and its application, 2013.01.
34. Hironobu Sugiyama, Takeshi Nanri, Performance Prediction Technology for Collective Communication Algorithm on Multi-Dimensional Mesh/Torus, International workshop on HPC, Krylov Subspace method and its application, 2013.01.
35. Matthew Livesey, James Francis Stack, Jr., Fumie Costen, Takeshi Nanri, Norimasa Nakashima, Seiji FUJINO, An Alternative Domain Decomposition Technique for CUDA-based 3D FDTD Methods, 9th European Radar Conference, 2012.11.
36. Matthew Livesey, James Francis Stack, Jr., Fumie Costen, Takeshi Nanri, Norimasa Nakashima, Seiji FUJINO, Impact of GPU Memory Access Patterns on FDTD, IEEE Antennas and Propagation Society International Symposium (APSURSI), 2012.07.
37. Seiji FUJINO, Takeshi Nanri, Kenichirou Kusaba, Balancing Communication and Execution Technique for Parallelized Sparse Matrix-Vector Multiplication, 4th International Conference on Future Computational Technologies and Applications, 2012.07.
38. FUKAZAWA Keiichiro, Takeshi Nanri, Effective Performance of Large-Scale MHD Simulation for Planetary Magnetosphere with Massively Parallel Computer, JSST2012 International Conference on Simulation Technology, 2012.07.
39. Takeshi Nanri, Motoyoshi Kurokawa, Efficient Runtime Algorithm Selection of Collective Communication with Topology-Based Performance Models, International Conference on Parallel and Distributed Processing Techniques and Applications, 2012.07.
40. Yoshiyuki Morie, Takeshi Nanri, Task Allocation Optimization for Neighboring Communication on Fat Tree, 14th IEEE International Conference on High Performance Computing and Communication, 2012.06.
41. FUKAZAWA Keiichiro, Takeshi Nanri, Performance of Large Scale MHD Simulation of Global Planetary Magnetosphere with Massively Parallel Scalar Type Supercomputer Including Post Processing, 14th IEEE International Conference on High Performance Computing and Communication, 2012.06.
42. Implementation and Performance Evaluation of Parallelized CG method with PAGME - Preconditioning Method on Hierarchical Parallel Computers.
43. Evaluation of Dynamic Algorithm Selection with Performance Prediction Models on Alltoall Operation.
44. Performance analysis of the CG method with parallelized PAGME.
45. Dependence on combination with number of processes and threads for com-
putation times of hybrid-parallel version of IDR(s) Method
.