Introduction to GPU computing with PyTorch
Friday, March 27, 2026 -
10:00 AM
Monday, March 23, 2026
Tuesday, March 24, 2026
Wednesday, March 25, 2026
Thursday, March 26, 2026
Friday, March 27, 2026
10:00 AM
Introduction to GPU Multiprocessing
Introduction to GPU Multiprocessing
10:00 AM - 10:15 AM
- GPGPU computing paradigm and typical application domains - Overview of CUDA and hardware-agnostic approaches
10:15 AM
Introduction to PyTorch
Introduction to PyTorch
10:15 AM - 11:45 AM
- Tensors: creation, initialisation and parameters - Aggregation and shape operations - Indexing, slicing and broadcasting, boolean and masked tensors - Matrix multiplication and elementwise math - Linear Algebra Using PyTorch
11:45 AM
Break
Break
11:45 AM - 12:15 PM
12:15 PM
GPU acceleration using PyTorch
GPU acceleration using PyTorch
12:15 PM - 1:45 PM
- Memory management in PyTorch - Comparing GPU vs CPU performance on linear algebra workloads - Motivation and basic principles of performance profiling - Profilers: setup, tracing and visualisation
1:45 PM
Break
Break
1:45 PM - 2:15 PM
2:15 PM
Custom GPU Kernels with Triton
Custom GPU Kernels with Triton
2:15 PM - 3:45 PM
- Motivation for writing custom kernels (performance, flexibility) - Overview of Triton and its programming model - Implementing basic kernels (e.g. vector operations, simple reductions) - Integration with PyTorch and comparison to built-in oper