Multi-GPU Programming Bootcamp
from
Monday, 28 September 2026 (12:00)
to
Wednesday, 30 September 2026 (13:30)
Monday, 28 September 2026
12:00
Cluster Dry Run Session (optional; please attend in case you need help with login)
Cluster Dry Run Session (optional; please attend in case you need help with login)
12:00 - 13:00
Tuesday, 29 September 2026
08:45
Join in
Join in
08:45 - 09:00
09:00
Welcome to the Multi-GPU Programming Bootcamp (Moderator)
Welcome to the Multi-GPU Programming Bootcamp (Moderator)
09:00 - 09:15
09:15
Introduction to Nsight Systems and GPU Computing (Lecture and Read-Only Lab)
Introduction to Nsight Systems and GPU Computing (Lecture and Read-Only Lab)
09:15 - 09:40
09:40
Overview of a Single GPU Code
Overview of a Single GPU Code
09:40 - 10:30
10:30
Break
Break
10:30 - 10:45
10:45
Parallelizing a Single GPU Code using CUDA memcpy, Streams, and an Introduction to Single Node Multi-GPU (P2P) (Lecture and Lab)
Parallelizing a Single GPU Code using CUDA memcpy, Streams, and an Introduction to Single Node Multi-GPU (P2P) (Lecture and Lab)
10:45 - 12:30
12:30
Lunch
Lunch
12:30 - 13:00
13:00
Introduction to Multi-Node Multi-GPU (GPU Direct) programming (Lecture and Lab)
Introduction to Multi-Node Multi-GPU (GPU Direct) programming (Lecture and Lab)
13:00 - 15:00
Wednesday, 30 September 2026
09:00
Join in
Join in
09:00 - 10:55
Introduction to NVIDIA Collectives Communications Library (NCCL) (Lecture and Lab)
10:55
Break
Break
10:55 - 11:10
11:10
Introduction to NVSHMEM (Lecture and Lab)
Introduction to NVSHMEM (Lecture and Lab)
11:10 - 13:00
13:00
Demo of the Nsight Systems Profiler Report for a Multi-Node Application
Demo of the Nsight Systems Profiler Report for a Multi-Node Application
13:00 - 13:30