Introduction to Supercomputing at JSC - Theory & Practice (Trainingskurs, online)
Ilya Zhukov, Jolanta Zjupa
(Kurs-Nr. 12520232 im Trainingsprogramm 2023 des Forschungszentrums Jülich)
Der Kurs findet als Online-Veranstaltung statt und richtet sich ausschließlich an Nutzerinnen und Nutzer der Jülicher Supercomputer. Der Link zur Online-Plattform wird nur den registrierten Teilnehmer:innen bekannt gegeben.
Inhalt:
Disclaimer: This course offers the opportunity to attend lectures selectively, based on individual needs and knowledge levels. Participants are not required to attend all sessions as some may cover advanced or basic material. While flexibility is offered, it is important to assess personal needs and choose sessions accordingly, as attending lectures out of order may result in knowledge gaps.
Research Centre Jülich provides cutting-edge high-performance computing resources to scientific groups and industry partners across Germany and Europe via the John von Neumann Institute for Computing. To help new users of JSC's supercomputers efficiently leverage their allocated resources, we offer an introductory course that covers system basics and best practices.
The course includes theoretical lectures held every afternoon from Monday to Thursday, and practical tutorials offered in the mornings from Tuesday to Thursday. The tutorials are based on the previous afternoon's lectures and allow articipants to put theory into practice. We cover a wide range of topics, starting from basic log-in procedures to intermediate-level techniques. Participants are free to choose the lectures that best match their needs and interests.
Topics covered include:
- User account management with the JuDoor portal
- System access via SSH, Jupyter, and UNICORE
- System configuration for JURECA and JUWELS
- File systems, I/O, and data management
- Software modules (compilers, MPI, math libraries, applications, debuggers, tools)
- Building software from source
- Submitting jobs via the resource manager
- Using GPUs
- Performance tuning
- Deep Learning
- Visualization
Contents level | in hours | in % |
---|---|---|
Beginner's contents: | 17.5 h | 80 % |
Intermediate contents: | 4.5 h | 20 % |
Advanced contents: | 0 h | 0 % |
Community-targeted contents: | 0 h | 0 % |
Voraussetzungen:
- Eigener Laptop mit Windows, GNU/Linux oder macOS
- Kenntnisse in C, C++ oder Fortran sind hilfreich
Agenda
The preliminary agenda can be found here.
Zielgruppe:
Neue Nutzerinnen und Nutzer der Supercomputer in Jülich
Sprache:
Der Kurs wird auf Englisch gehalten.
Dauer:
3,5 Tage
Zeit:
20-23 November 2023, startet am Montag um 13:00 Uhr
Tuesday to Thursday: 09:00-17:00 Uhr
Ort:
Online
Teilnehmerzahl:
maximal 100
Referentinnen und Referenten:
Mitarbeitende des JSC, Firmenangehörige von Atos, Intel und ParTec
Kontakt:
- Institute for Advanced Simulation (IAS)
- Jülich Supercomputing Centre (JSC)
Raum 323
- Institute for Advanced Simulation (IAS)
- Jülich Supercomputing Centre (JSC)
Raum 324
Kursmaterial
Titel
Hinweis
Link
Erscheinungsdatum
Title | Video |
---|---|
Welcome and Introduction to JSC | |
HPC in a Nutshell | |
User Portal and Job Monitoring | |
JUST - Jülich Storage Cluster | |
JSC Systems — JUWELS, JURECA & JUSUF | |
HPC Software — Modules, Libraries & Software | |
Batch usage on JSC | |
JupyterLab - Supercomputing in your Browser | |
Proper Pinning Prevents Pretty Poor Performance | |
Uniform Resource Access - Compute and Cloud Resources at JSC | |
GPU accelerators at JSC | |
Deep Learning on Supercomputers | |
HPC Software - Debuggers and Performance Analysis Tools | |
Hands-On: Remote Visualization | |
Intel tuning for JUWELS and JURECA | |
JUWELS and JURECA Tuning for the platform | |
Evolution of the Sequana System Architecture |