Introduction to Scientific High-Performance Computing

Tuesday 26 Nov 2024, 10:00 → 15:00 Europe/Warsaw

online

Maciej Czuchry (ACC Cyfronet AGH), Oskar Klimas (ACK Cyfronet AGH), Klemens Noga (ACC Cyfronet AGH)

The "Introduction to Scientific High-Performance Computing" course will deliver essential information for users who want to start working with the HPC system available at ACC Cyfronet AGH.

Agenda

• Introduction to HPC systems
  • login and data transfer
  • basics on how clusters/supercomputers work
• Basics of environmental variables in bash and their usage on HPC clusters
• Available computational and storage resources of Ares and Athena supercomputers
• Working with the queuing system Slurm Workload Manager 
  • batch scripts (with examples)
    • sequential, parallel and distributed computations
  • interactive jobs
  • efficient usage of the Slurm queuing system
• Scientific software environment management (Modules/Lmod)
• Python virtual environment management
• Getting user support through Cyfronet's Helpdesk and best practices

Duration

5 hours

Requirements

Each user should have a PLGrid account, active affiliation and access to the Ares cluster. Accounts and access to supercomputers could be obtained in Portal PLGrid. If you do not have an active affiliation, please get in touch with organisers at training@cyfronet.pl.

Venue

Training will be conducted via Zoom through the link provided for registered participants.

Language

English/Polish - dependent on the participants.

Level

beginner

10:00 → 10:15 HPC Intro

Introduction to HPC systems, logging and transfering files

Speaker: Oskar Klimas (ACK Cyfronet AGH)

10:15 → 10:45 What is a supercomputer? (PC vs cluster)

Basics on how clusters/supercomputers work

Speaker: Oskar Klimas (ACK Cyfronet AGH)

10:45 → 11:00 Environemental variables

Basics of environmental variables in bash and their usage on HPC clusters

Speaker: Oskar Klimas (ACK Cyfronet AGH)

11:00 → 11:30 File systems on Cyfronet clusters and how to use them

Available computational and storage resources of Ares, Athena and Helios supercomputers

Speaker: Oskar Klimas (ACK Cyfronet AGH)

11:30 → 12:00 Basics of quening system and batch scripts

Working with the queuing system Slurm Workload Manager
batch scripts (with examples)
sequential, parallel and distributed computations

Speaker: Oskar Klimas (ACK Cyfronet AGH)

12:00 → 12:30 Lunch break

12:30 → 13:00 Scientific software management (modules)

Scientific software environment management (Modules/Lmod)

Speaker: Oskar Klimas (ACK Cyfronet AGH)

13:00 → 13:30 Python virtual environment

Python virtual environment management

Speaker: Oskar Klimas (ACK Cyfronet AGH)

13:30 → 14:00 Scripting for quening system

Working with the queuing system Slurm Workload Manager
batch scripts (with examples)
sequential, parallel and distributed computations

Speaker: Oskar Klimas (ACK Cyfronet AGH)

14:00 → 14:30 HANDS-ON - scripting with all the things you've learned

Working with the queuing system Slurm Workload Manager
batch scripts (with examples)
sequential, parallel and distributed computations

Speaker: Oskar Klimas (ACK Cyfronet AGH)

14:30 → 14:45 Modes of work with Slurm

interactive jobs, efficient usage of the Slurm queuing system

Speaker: Oskar Klimas (ACK Cyfronet AGH)

14:45 → 15:00 How to get support with a problem from Helpdesk

Getting user support through Cyfronet's Helpdesk and best practices

Speaker: Oskar Klimas (ACK Cyfronet AGH)