OpenMP with AMD GPU Hackathon – Itinerary

The hackathon is a 3-day event where most of the time will be spent on enhancing your project codes.

A 3-day pre-event training for all those that would like to attend will be as follows

Pre-event training 20th March

Time	Topic
13:00 – 13:10	AMD Presentation Intro
13:10 – 13:20	Brief Introduction to AMD Architecture
13:20 – 14:10	Getting Started with OpenMP® Offload Applications on AMD Accelerators
14:10 – 14:40	OpenMP® exercises
14:40 – 14:50	Break
14:50 – 15:00	OpenACC
15:00 – 16:00	Advanced OpenMP® offloading
16:00 – 16:30	Advanced OpenMP® Exercises
16:30 – 16:40	Break
16:40 – 16:55	Mixing HIP and OpenMP®
16:55 – 17:00	Wrapup

Pre-event training 21st March

Time	Topic
13:00 – 13:30	AMD Communication Fabrics and GPU-Aware MPI
13:30 – 13:45	GPU Aware Exercises
13:45 – 14:00	Break
14:00 – 14:40	AMD Node Memory Model
14:40 – 15:00	Memory Model Exercises
15:00 – 15:15	Break
15:15 – 16:15	Affinity — Placement, Ordering and Binding
16:15 – 16:40	Affinity exercises
16:40 – 16:55	Infinity Hub/HPC Community
16:55 – 17:00	Wrapup

Pre-event training 22nd March

13:00 – 13:20	Introduction to Rocprof
13:20 – 13:35	Rocprof Exercises
13:35 – 14:00	Introduction to Omnitrace
14:00 – 14:15	Omnitrace Exercises
14:15 – 14:25	Break
14:25 – 15:25	Roofline
15:25 – 15:55	Introduction to Omniperf
15:55 – 16:10	Exercises – Omniperf
16:10 – 16:20	Break

Extra

Intro to AMD Instinct Accelerators.mp4