This half-day course is dedicated to learning how to efficiently use the GPU accelerated part of Karolina for Deep and Machine Learning. But it can be aplied to any multi-GPU system.
Short introduction of the Karolina supercomputer How to access the Karolina GPU nodes First login Computing environment and available software libraries and tools HPC resources allocation, PBS Scratch and Project storages Special tools (Nodes availability overview, ...) Introduction to Data Parallel Deep Learning with Horovod Multi-node/-GPU aware Data Processing Pipelines Demonstration of Multi-node/-GPU Examples using Tensorflow Multi-node/-GPU Machine Learning with scikit-learn Efficient execution of a large number of small tasks transparently over HPC schedulers (SLURM/PBS) using HyperQueue
Access to Karolina's GPU accelerated part Efficient multi-GPU and multi-node execution of Deep and Machine Learning frameworks Introduction to HyperQueue
Digital Innovation Hubs
IT4Innovations National Supercomputing Center