Scalability - smart use of HPC resources
online
The training focuses on analyzing the scalability tests.
Scalability tests allow to tune the amount of resources for the job being run. During this training you will learn what scalability tests are and how to interpret their results. Then how to apply that knowledge to your calculations.
Agenda
- Intranode and perfect scaling, and how it is unattainable in real calculations,
- Multinode scaling and parallelization overhead,
- CPU vs GPU power and multiple GPU scaling,
- Real life case: specific Ares to Athena transfer issue analyzed
Duration
2 hours
Target audience
Training is intended for users who started running their own calculations and need to learn how to give the right amount of resources to the job.
Requirements
We recommend doing this training after completing Running Calculations the HPC way training provided at ACK Cyfronet AGH.
Technical prerequisites:
- PLGrid account at https://portal.plgrid.pl/,
- an active affiliation at https://portal.plgrid.pl/users/affiliations,
- access to Ares supercomputer at https://portal.plgrid.pl/services/103,
- access to JupyerHubHPC at https://portal.plgrid.pl/services/110,
- membership in the plggtraining team - the activation link will be sent after registration.
The above prerequisites are required to actively participate in the hands-on exercises.
In case of no active affiliation or PLGrid account, please contact the organisers at training@cyfronet.pl.
Participants' computers should be equipped with software that allows SSH connections. On Windows, it is recommended to use Windows Terminal after checking if OpenSSH is installed. We advise against using Putty.
Venue
The workshop will be conducted online on the Zoom platform. The meeting link will be sent to registered participants.
Language
English/Polish - dependent on the participants.
Level
Intermediate
ACC Cyfronet AGH
