legioん

本文へ toggle navigation About CCS column1 Message from the Director Mission Chronology Organization column2 Scientists Honorary Fellowship Faculty and Research Fellow Division of Particle Physics Division of Astrophysics Division of Nuclear Physics Division of Quantum Condensed Matter Physics Division of Life Sciences Division of Global Environmental Science Division of High Performance Computing Systems Division of Computational Informatics Collaborative Fellow (Other Organization) column3 Careers Findings Research Papers and Conference Presentations Research Papers and Conference Presentations FY2009 External Review External Review (2008-2013) External Review(2007) SC Exhibit Posters Research Activities column1 Project Office for Exascale Computational Sciences Project Office for Exascale Computing System Research and Development Project Office for HPCI Strategic Program Project Office for AI & Big Data Promotion Office for Computational Sciences Project Office for High Performance Computing System Development and Management Liaison Office for Multidisciplinary Computational Sciences Project Office for Society-Academia collaboration in computational sciences Bureau of Public Relations and Strategy Department of Computational Medical Science Research Divisions column1 Division of Particle Physics Division of Astrophysics Division of Nuclear Physics Division of Quantum Condensed Matter Physics Division of Life Sciences Biological Function and Information Group Molecular Evolution Group column2 Division of Global Environmental Science Division of High Performance Computing Systems Division of Computational Informatics Database Group Computational Media Group Projects column1 Joint Center for Advanced High Performance Computing (JCAHPC) column2 Published Codes Databases column3 Past Projects Supercomputers Open Seminars column1 Symposium Colloquium Workshop Use Computers column1 Multidisciplinary Cooperative Research Program (MCRP) MCRP application Guideline for acknowledgement MCRP: Procedure after adoption Public Relations column1 Press Releases Brochure Logo Mark Gallery Contact Japanese Contact Careers RSS Traffic access font size: medium large About CCS column1 Message from the Director Mission Chronology Organization column2 Scientists Honorary Fellowship Faculty and Research Fellow Division of Particle Physics Division of Astrophysics Division of Nuclear Physics Division of Quantum Condensed Matter Physics Division of Life Sciences Division of Global Environmental Science Division of High Performance Computing Systems Division of Computational Informatics Collaborative Fellow (Other Organization) column3 Careers Findings Research Papers and Conference Presentations Research Papers and Conference Presentations FY2009 External Review External Review (2008-2013) External Review(2007) SC Exhibit Posters Research Activities column1 Project Office for Exascale Computational Sciences Project Office for Exascale Computing System Research and Development Project Office for HPCI Strategic Program Project Office for AI & Big Data Promotion Office for Computational Sciences Project Office for High Performance Computing System Development and Management Liaison Office for Multidisciplinary Computational Sciences Project Office for Society-Academia collaboration in computational sciences Bureau of Public Relations and Strategy Department of Computational Medical Science Research Divisions column1 Division of Particle Physics Division of Astrophysics Division of Nuclear Physics Division of Quantum Condensed Matter Physics Division of Life Sciences Biological Function and Information Group Molecular Evolution Group column2 Division of Global Environmental Science Division of High Performance Computing Systems Division of Computational Informatics Database Group Computational Media Group Projects column1 Joint Center for Advanced High Performance Computing (JCAHPC) column2 Published Codes Databases column3 Past Projects Supercomputers Open Seminars column1 Symposium Colloquium Workshop Use Computers column1 Multidisciplinary Cooperative Research Program (MCRP) MCRP application Guideline for acknowledgement MCRP: Procedure after adoption Public Relations column1 Press Releases Brochure Logo Mark Gallery Contact colloquium English Top トップページ Seminar colloquium 137th Colloquium of the Center for Computational Sciences colloquium 137th Colloquium of the Center for Computational Sciences 137th Colloquium Date: 15 January 2024Time: 14:00-15:30Venue: Center for Computational Sciences, Workshop Room Language: English 14:00-14:45 Challenges for Extreme Scale Sparse Matrix computing 【Abstract】   Prof. Serge G. Petiton (University of Lille and CNRS, France) 14:45-15:30 High Performance Learning with the Unite and Conquer Approach 【Abstract】   Prof. Nahid Emad (University of Paris-Saclay/Versailles, France) Abstract1: Exascale machines are now available, based on several different arithmetic (from 64-bit to 16- 32 bit arithmetics, including mixed versions and some that are no longer IEEE compliant) and using different architectures (with network-on-chip processors and/or with accelerators). Brain-scale applications, from machine learning and AI for example, manipulate huge graphs that lead to very sparse non-symmetric linear algebra problems. Moreover, those supercomputers have been designed primarily for computational science, mainly numerical simulations, not for machine learning and AI. New applications that are maturing after the convergence of big data and HPC to machine learning and AI would probably generate post- exascale computing that will redefine some programming and application development paradigms. End-users and scientists have to face a lot of challenge associated to these evolutions and the increasing size of the data. In this talk, after a short description of some recent evolutions having important impacts on our results, in particular about programming paradigms. I present some results obtained on the still #1 supercomputer of the HPCG list, Fugaku, for sequences of sparse matrix products, with respect to the sparsity and the size of the matrices, on the one hand, and to the number of process and nodes, on the other hand. Then, I introduce two opensource generators of very large data, allowing to evaluate several methods using very large graph-sparse matrices as data sets for several application evaluations.   Abstract 2： The ever-increasing production of data requires new methodological and technological approaches to meet the challenge of their effective analyses. We highlight the omnipresence of certain linear algebra methods such as the eigenvalue problem or more generally the singular value decomposition in machine learning techniques. A new machine learning approach based on Unite and Conquer methods, used in linear algebra, will be presented. The important characteristics of this intrinsically parallel and scalable technique make it very well suited to multi-level and heterogeneous parallel and/or distributed architectures. We also highlight the very strong interactions between machine and deep learning methods and sparse linear algebra and present a promising approach in this area. Experimental results demonstrating the interest of these approaches for efficient data analysis in the case of clustering, cybersecurity and road traffic simulation will be presented. Coordinator : Boku Taisuke Facebook で共有するにはクリックしてください (新しいウィンドウで開きます)クリックして Twitter で共有 (新しいウィンドウで開きます)« 136th Colloquium of the Center for Computational Sciences[Cancel] 140th Colloquium of the Center for Computational Sciences » About CCSFindings SC Exhibit Posters External Review External Review(2007) External Review (2008-2013) Parallel Tracks for Divisions External Review Photo Gallery Research Papers and Conference Presentations Research Papers and Conference Presentations FY2009 Scientists Collaborative Fellow (Other Organization) Faculty and Research Fellow Division of Computational Informatics Division of High Performance Computing Systems Division of Global Environmental Science Division of Life Sciences Division of Quantum Condensed Matter Physics Division of Nuclear Physics Division of Astrophysics Division of Particle Physics Honorary Fellowship Organization Chronology Mission Message from the Director Research DivisionsDivision of Computational Informatics Computational Media Group Database Group Division of High Performance Computing Systems Division of Global Environmental Science Division of Life Sciences Molecular Evolution Group Biological Function and Information Group Division of Quantum Condensed Matter Physics Division of Nuclear Physics Division of Astrophysics Division of Particle Physics Research ActivitiesDepartment of Computational Medical Science Bureau of Public Relations and Strategy Project Office for Society-Academia collaboration in computational sciences Liaison Office for Multidisciplinary Computational Sciences Project Office for High Performance Computing System Development and Management Promotion Office for Computational Sciences Project Office for HPCI Strategic Program Project Office for AI & Big Data Project Office for Exascale Computing System Research and Development Project Office for Exascale Computational Sciences ProjectsHA-PACS History of PACS/PAX Series Progress and Achievements HA-PACS Base Cluster Exascale Computing System Development Exascale Computational Sciences Organization of HA-PACS Project Project Objective Past Projects Joint Center for Advanced High Performance Computing (JCAHPC) Published Codes QCD (Quantum Chromo Dynamics) Databases Supercomputers Open SeminarsColloquium Workshop Symposium Use ComputersCCS security policy FAQ Guideline for acknowledgement MCRP: Procedure after adoption MCRP application Multidisciplinary Cooperative Research Program (MCRP) Public RelationsOld Web Site Contact Gallery Logo Mark Brochure Press Releases About this site Center for Computational Sciences, University of Tsukuba1-1-1 Tennodai, Tsukuba, Ibaraki 305-8577 Japan +81-29-853-6487 （CCS Office） © 2016 Center for Computational Sciences, University of Tsukuba.