5. Call to Action

The following list suggests things that can be done today or as part of a GPUDirect Storage implementation.

• Choose to be part of the GPU storage platform of the future.

• Enable your application by fully porting it to the GPU, so that the IO is directly between GPU memory and storage.

• Use interfaces that make explicit transfers: use cuFile APIs directly or via a framework layer that is already enabled to use cuFile APIs.

• Use cuFile APIs for any type of memory allocation, including memory on the CPU, so that a consistent set of interfaces is used for storage throughout your application. Make use of the GPUDirect Storage Best Practices Guide to select and apply the best APIs to use.

• Choose and use distributed file systems or distributed block systems that are enabled with GPUDirect Storage.