GPU
A processor built for massive parallelism, which is what neural network maths needs.
Overview
A GPU is a processor with thousands of simple cores optimised for doing the same operation across large arrays at once. It solves the problem that training and serving a neural network is almost entirely matrix multiplication, which is exactly that pattern, while a CPU spends its few complex cores poorly on it. It exists as the reason deep learning became practical: the hardware already existed for graphics, and it happened to fit the maths.
Technology Graph
How GPU connects to the rest of the stack.
Uses and depends on
- DeepSpeed framework Frameworks DeepSpeed depends on GPU High ✓ source
- vLLM tool Infrastructure vLLM depends on GPU High ✓ source
Applied to
- Deep Learning concept Algorithms GPU used for Deep Learning High ✓ source
- NVIDIA CUDA platform Hardware NVIDIA CUDA used for GPU High ✓ source
Classification
- Tensor Core hardware Hardware Tensor Core part of GPU High ✓ source
Alternatives
Comparisons
Side-by-side breakdowns with the alternatives above.