Titna X or Titan V (Volta Architecture) for low latency GPUDirect data transfer btw. FPGA and GPU Memory

Hallo Collegues,

I intend to use the Titan V / or Titan X GPU card for low latency video data transfer from a Ximea PCIe Camera into GPU-Memory without usage of CPU RAM.

I can not find information, if the Titan V or Titan X GPU Cards does support the GPU Direct transfer on Windows or Linux.

On every documentation, I only read, that the Tesla, Quandro or even Pascal architectures does support the GPUDirect transfers, but nowhere is mentione somethink about Volta architecture of Titan V.

I’m wondering if the Volta GPU Architecture of Titan V, as a newer design does support already used GPUDirect or P2P Technologies, which was developed and standard in former Architectures since many years, like you can see in the Nvidia roadmap at SGEMM/W 2018.

Does the Titan V Volta Architecture support GPUDirect technology, or not ?

Where can I find a list of GPU Cards or Embedded Modules supporting GPUDirect ?

Thank you in advance.

I’ve also requested this but even support wasn’t able to help me. The best I could find is, if it says Tesla (since Kepler?), Yes, if anything else, No. Please correct me if other (cheaper) cards are also supported.

This seems to be a very frequently asked question:
https://devtalk.nvidia.com/default/topic/716091/?comment=5267361

I’m also looking for more information on how to support GPUDirect RDMA/Async on the NIC side. What should be implemented etc. if anyone has information it would be much appreciated:
https://devtalk.nvidia.com/default/topic/1042508/other-libraries/gpudirect-rdma-async-for-dl-acceleration-mpi-nccl-/

Sorry to hear about your issue. This board is focused on the use of the NVIDIA GPU Cloud containers so it isn’t the best place to address your question. A board like https://devtalk.nvidia.com/default/board/236/gpu-hardware/ might be more productive.