CUDA
CUDA (Compute Unified Device Architecture) is a parallel computing platform and programming model developed by NVIDIA that enables developers to use NVIDIA GPUs for general-purpose processing, vastly accelerating computationally intensive applications. It provides a software layer that gives…
Key facts
- First appeared
- 2006
- Category
- technology
- Problem solved
- CUDA was created to address the significant challenge of efficiently utilizing the immense parallel processing power of Graphics Processing Units (GPUs) for tasks beyond traditional graphics rendering. Before CUDA, leveraging GPUs for general-purpose computation (GPGPU) was an arduous process, requiring developers to creatively map non-graphics problems onto graphics APIs like OpenGL or DirectX, often forcing them into rendering metaphors that were unnatural and difficult to program. This limited the accessibility and efficiency of GPGPU for scientific computing, simulations, and data processing.
- Platforms
- Windows (x86-64), macOS (x86-64, prior to recent ARM transition), Linux (x86-64, ARM64), NVIDIA GPUs (GeForce, Quadro, Tesla, Jetson)
Related technologies
Notable users
- Tesla
- Microsoft
- Academic Research Institutions
- OpenAI
- IBM
- Lawrence Livermore National Laboratory (LLNL)
- Meta (Facebook)
- Amazon Web Services (AWS)
- Oak Ridge National Laboratory (ORNL)