Allows CUDA developers to deploy kernels requiring more global GPU memory than is available by transparently mapping pinned CPU memory into the GPU memory map. Works behind the scenes and is easily added to existing projects with minimal effort.
- Use Plan files to determine memory allocation locations at runtime.
- Plans can be automatically determined by built-in tuners
- Includes built-in Genetic and Exhaustive tuners for Plan generation
- Minimally invasive. Programmers only need to add 3 or 4 lines of code.
Be the first to post a review of libcuzmem!