The KAUST Supercomputing Core Lab and KAUST Visualization Core Lab are organizing a training session to onboard KAUST users to the Ibex Cluster for running Data Science and AI workloads. The event is designed to train the current and future users of Ibex to apply best practices to their workloads and understand policies through hands-on exercises and demos.Who should attend AI model developers wanting to access HPC resources e.g. GPUs and multicore CPUsLearning outcomes After attending, the participants will be able to:
· Request computational resources on Ibex appropriately and responsibly
· Be able to interact with managed and self-serviced software environments on Ibex
· Be able to run interactive and batch jobs on the allocated computational resources
· Understand the storage on Ibex and the related best practices and policies
Agenda TimeTopic 08:00 Coffee 08:30 Computational resources on Ibex for Data science and AI workloads 09:15 Model development lifecycle on Ibex 11:00 Portable software environments with containers 12:00 Distributed training jobs on Ibex