Access to LUMI

Course/Event Essentials

Event/Course Start
Event/Course End
Event/Course Format
Mixed
Live (synchronous)
Primary Event/Course URL
https://events.it4i.cz/event/160/

Venue Information

Country: Czech Republic
Venue Details: Click here

Training Content and Scope

Scientific Domain
Not scientific domain specific
Technical Domain
General HPC systems and architectures
Level of Instruction
Beginner
Intermediate
Advanced
Sector of the Target Audience
Research and Academia
Industry
Public Sector
HPC Profile of Target Audience
Application Users
Application Developers
Data Scientists
Language of Instruction
English

Other Information

Organiser
IT4Innovations National Supercomputing Center
Supporting Project(s)
EuroCC/CASTIEL
Event/Course Description

This course is dedicated to the supercomputer LUMI. EuroHPC Joint Undertaking’s LUMI supercomputer reached the third spot on the Top500 list of the world’s fastest supercomputers released at the SC22 conference in Dallas, USA, on 14 November 2022. LUMI reached a measured High-Performance Linpack (HPL) performance of 309 petaflops. This makes LUMI the fastest supercomputer in Europe.

Participants will learn about its architecture and key parameters, how to access LUMI resources, how to run jobs, and how to use the GPU accelerated partition for AI applications. 

The course will be delivered in hybrid mode. You can attend the course at IT4Innovations in Ostrava or online. 

Tutors

Mgr. Branislav Jansík, Ph.D. - Supercomputing Services Director of IT4Innovations

Jan Vicherek - LUMI Support Team, IT4Innovations

Georg Zitzlsberger - trainer and researcher at the Advanced Data Analysis and Simulations Lab, IT4Innovations

Detailed agenda

9:00 – 9:30 About LUMI (B. Jansík) 30 mins

9:30 – 10:15 Introduction of LUMI supercomputer (J. Vicherek) 45 mins

  • Architecture and key parameters
  • LUMI-C partition
  • LUMI-G partition
  • Scratch and Project storage

 10:15 – 10:30 Break

 10:30 – 11:30 How to access the LUMI supercomputer (B. Jansík + J. Vicherek) 30 mins + 30 mins

  • Access mechanisms
  • How to submit an application for computational resources
  • Evaluation of application
  • Account creation request
  • First login to the LUMI (guided examples)
  • How to run jobs (guided examples)

 11:30 – 12:30 Lunch Break

 12:30 – 13:30 The use of LUMI supercomputer (J. Vicherek) 60 mins

  • Computing environment and available software libraries and tools 
  • HPC resources allocation, SLURM (guided examples)
  • Scratch and Project storages (guided examples)

13:30 – 14:30 Efficient multi-GPU and multi-node execution of Deep Learning frameworks (G. Zitzlsberger) 60 mins

  • Introduction to Data Parallel Deep Learning with Horovod
  • Multi-node/-GPU aware Data Processing Pipelines
  • Demonstration of Multi-node/-GPU Examples using Tensorflow/Keras

 14:30 – 15:00 Q&A and closing