CUDA general-purpose GPU computing
CUDA, computación de propósito general en GPU
- Date
- 2007 · confidence high
- Place
- Nvidia, Santa Clara, California, United States
- Credited to
- Ian Buck and Nvidia
- Field
- Knowledge & symbols
What it is
Programming platform that lets general scientific code run on the thousands of cores of a GPU. It turned the graphics card into AI's supercomputer.
What it led to
- The deep-learning revolution from 2012 onwards
- Nvidia's competitive moat and the shortage of AI chips
- Scientific simulation and cryptocurrency mining on consumer hardware