TOP    Events & Outreach    Schools    RIKEN International HPC Summer School 2022 - Toward Society 5.0 -

This school is "hybridonline".
("online" for non-residents of Japan and “on-site” for Japanese residents.)

RIKEN Center for Computational Science (R-CCS) will hold an hybrid online summer school in September 2022.
This program is designed to provide young researchers in computational science with an opportunity to learn programming techniques for parallel computers and to help them become leaders in the field in the age of Society 5.0. Scientists from R-CCS will provide lectures, and the supercomputer Fugaku will be used for hands-on training.
This program will be provided in English.

This year, the program will be renewed as the one with 5-days. First 4 days are focusing on parallel Finite Element Method (FEM). FEM is widely used for solving various types of applications in science and engineering. This part includes fundamental issues related to FEM, parallel programming using MPI, OpenMP and numerical libraries, such as PETSc. Participants will gain knowledge and insights into scientific computing through hands-on sessions using Fugaku.

On the final day, we will deepen your understanding of numerical libraries by practicing PETSc applications in the morning. We will have a hands-on lecture on Deep Neural Networks in the afternoon, demonstrating a simple image categorization and bridge to the higher-level AI frameworks and toolkits.

Details
Date Mon, Sep 12 - Fri, Sep 16, 2022
City Online
Place

Online

Organizer

RIKEN Center for Computational Science

Sponsorship

Introduction of Lecturers

Administrative Office

Collaborations Group
RIKEN Computational Science Promotion Office

Qualifications

Eligible individuals are as follows.

  • Students who are enrolled in a Ph.D. program, post-doctoral scientists and early career scientists affiliated with universities or research institutions.
  • Scientists and engineers affiliated with private companies.

Please understand that individuals who do not meet requirements shown below may be declined to participate in the program.

  • Ideal participants have programming experience using C or Fortran, have used Linux.
  • (at least can use an editor on Linux), and have basic parallel programing experience.
  • Please be aware that online participants need an internet environment as follows.
    -Network bandwidth with a stable connection when using video and audio for the teleconference system such as Zoom.
    -Possible to connect remotely to Fugaku by using Secure Shell.

Program

Mon, Sep 12(JST)
Time Instructor Program/Materials
09:00 - 12:00 Nakajima Introduction, 1D FEM
13:30-17:00 Nakajima 3D FEM
17:00-18:00 Imamura Fugaku Virtual Tour, Overview of Fugaku

Tue, Sep 13(JST)
Time Instructor Program/Materials
09:00 - 12:00 Nakajima,
Terao
Road to Parallel FEM, Introduction to MPI(1/5)
13:30-14:30 Nakajima,
Terao
Introduction to MPI(2/5)
14:30-15:30 Terao Exercise(Optional)
15:30-18:00 Nakajima,
Terao
Introduction to MPI(3/5)

Wed, Sep 14(JST)
Time Instructor Program/Materials
09:00 - 10:30 Nakajima,
Terao
Introduction to MPI(4/5)
10:30-12:00 Terao Exercise(Optional)
13:30-15:00 Terao Exercise(Optional)
15:00-16:30 Nakajima,
Terao
Introduction to MPI(5/5)
16:30-18:00 Nakajima,
Terao
Parallel FEM (1/2)

Thu, Sep 15(JST)
Time Instructor Program/Materials
09:00 - 12:00 Nakajima,
Terao
Parallel FEM(2/2)
13:30-15:00 Nakajima,
Terao
Parallel Visualization
15:00-18:00 Nakajima,
Terao
OpenMP/MPI Hybrid

Fri, Sep 16(JST)
Time Instructor Program/Materials
09:00 - 12:00 Terao,
Nakajima
PETSc for Parallel FEM
13:30-17:30 Imamura,
Terao
Introduction to Machine Learning and Neral Network
17:30- Closing

Application for participation

  • Maximum number of participants: About 40. Applications will be accepted on a first-come, first-served basis. However, necessary adjustment may be made to avoid having a large number of participants from one university or institution. We may also decline applicants who do not meet the requirements listed in 5 above.
  • Application deadline: 5 pm, July 22, 2022 (JST) Closed
  • To apply for the summer school, click here (online application form).The webpage will open in a new tab.
  • After the application period ends, the summer school secretariat will notify all the applicants whether they are accepted for the program, and inform them regarding documents to submit.
  • In principle, participants must attend all sessions of the summer school. For partial participation, please consult with the R-CCS Computational Science Promotion Office.
  • (Only for Japanese residents )RIKEN will bear travel and accommodation expenses to full-time students who will participate in all sessions (part-time students with full-time jobs are excluded). The expenses will be paid based on the RIKEN’s regulations via bank transfer after the school ends.
    (There will be no payment of expenses if you choose online. If you wish to participate in the school on-site, you may not change your participation to online participation for your personal reasons.)

Note: Handling of personal information
Personal information submitted will be used only for RIKEN to inform participants of the school schedule and changes or to collect materials related to the school operations. Personal information will not be used for other purposes.

*If you are a resident of the European Economic Area (the EU zone), you are required to submit the GDPR consent form (PDF 169KB)新しいタブでPDFが開きます with your signature.
You will find detailed information about General Data Protection Regulation (GDPR) in the website below:
European Commision Data Protection Rules新しいタブでPDFが開きます

*School participants must be aware of "the Regulations for Use of the Supercomputer Fugaku (PDF 42KB) "新しいタブでPDFが開きます
Further details will be informed once participants are set.

Inquiries

RIKEN Computational Science Plannning Officen7-1-26, Minatojima-minamimachi, Chuo-ku, Kobe, 650-0047, JAPAN

  • mail addressrenkei_hrd[at]ml.riken.jp*
  • phone number078-940-5796
  • Fax number078-304-4956
  • *Please change [at] to @ when sending email

(Oct 7, 2022)