nvCOMP is a CUDA library that features generic compression interfaces to enable developers to use high-performance GPU compressors and decompressors in their applications. Example benchmarking results and a brief description of each algorithm are available on the nvCOMP Developer Page.
This folder demonstrates the nvCOMP library API usage alongside our performance benchmarks.
More information can be found about the examples here and about the benchmarks here
- Pascal (SM 6.x)
- Volta (SM 7.0, and SM 7.2)
- Turing (SM 7.5)
- Ampere (SM 8.0, SM 8.6, and SM 8.7)
- Ada Lovelace (SM 8.9)
- Hopper (SM 9.0)
More information can be found about the architectures and compute capabilities on the official NVIDIA website here.
Linux, Windows
x86_64, aarch64
- A Linux or Windows system with recent NVIDIA drivers.
- CUDA 11.8 or CUDA 12.5 toolkit.
- CMake 3.18 or above
- gcc-8 or above (with C++11 support)
- The latest nvCOMP release from the NVIDIA developer zone.