Today Nvidia announced the general availability of the CUDA 8 toolkit for GPU developers.
A crucial goal for CUDA 8 is to provide support for the powerful new Pascal architecture, the first incarnation of which was launched at GTC 2016: Tesla P100,” said Nvidia’s Mark Harris in a blog post. “One of NVIDIA’s goals is to support CUDA across the entire NVIDIA platform, so CUDA 8 supports all new Pascal GPUs, including Tesla P100, P40, and P4, as well as NVIDIA Titan X, and Pascal-based GeForce, Quadro, and DrivePX GPUs.”
- Support for the Pascal GPU architecture, including the new Tesla P100, P40, and P4 accelerators
- New Unified Memory capabilities
- Native FP16 and INT8 computation for deep learning and other workloads
- The new nvGRAPH GPU-Accelerated Graph Analytics library
- Powerful new profiling capabilities
- Improved compiler performance and heterogeneous lambda support
- Expanded developer platform support including Microsoft Visual Studio 2015 (updates 2 and 3) and GCC 5.4 (Ubuntu 16.04)
Read the Full Story on Cuda 8 capabilities and performance enhancements.