Best practices in High-Perfomance Computing (sciCORE cluster)

This course will be streamed only for the registered participants. Registered participants will receive specific information directly from the respective course’s organizers.

Overview

This one-day course is a follow-up to the sciCORE course "Introduction to HPC (sciCORE cluster)". We will focus on different aspects related to the everyday usage of HPC facilities in order to enable an efficient and productive framework to develop your research. These include (among other topics) advanced SLURM usage, I/O management, and profiling.

Audience

This course is aimed at researchers that want to deepen their knowledge of the usage of HPC facilities. In particular, those that already took the introductory course to high-performance computing (although it is not a prerequisite).

Learning objectives

At the end of the course, the participants are expected to:

• Have a good understanding of advanced capabilities of SLURM. Including dependencies, parallelism, and monitoring
  

• Know good practices regarding the data management and its relation with the file system. This will include how to work with workflows and the usage of Labkey
  

• Learn how to use the sciCORE infrastructure for scientific applications (including hands-on exercises)
  

• Know the basics of profiling code and the impact of scaling
  

• How to speed up Python code
  

• The basics of parallelize in R
  

• How to manage errors
  

• Have an overview of online production tools such us Git, Slack, and Overleaf
  

• Have some advanced tips for the command line
  

Prerequisites

Knowledge / competencies

Knowledge on Linux operating system and the Slurm scheduler.

Application

The course is free of charge, but registration is mandatory.

You will be informed by email of your registration confirmation.

Venue and Time

The course will be hosted online.

The course will start at 9:00 and end around 17:00.

Precise information will be provided to the participants on due time.

Additional information

You are welcome to register to the SIB courses mailing list to be informed of all future courses and workshops, as well as all important deadlines using the form here.

For more information, please contact training@sib.swiss.