How can I make tensorflow run on a GPU with capability 2.x?

by

Recent GPU versions of tensorflow require compute capability 3.5 or higher (and use cuDNN to access the GPU.

cuDNN also requires a GPU of cc3.0 or higher:

cuDNN is supported on Windows, Linux and MacOS systems with Pascal, Kepler, Maxwell, Tegra K1 or Tegra X1 GPUs.

  • Kepler = cc3.x
  • Maxwell = cc5.x
  • Pascal = cc6.x
  • TK1 = cc3.2
  • TX1 = cc5.3

Fermi GPUs (cc2.0, cc2.1) are not supported by cuDNN.

Older GPUs (e.g. compute capability 1.x) are also not supported by cuDNN.

Note that there has never been either a version of cuDNN or any version of TF that officially supported NVIDIA GPUs less than cc3.0. The initial version of cuDNN started out by requiring cc3.0 GPUs, and the initial version of TF started out by requiring cc3.0 GPUs.

More Related Contents:

  1. Which TensorFlow and CUDA version combinations are compatible?
  2. Is it still necessary to install CUDA before using the conda tensorflow-gpu package?
  3. How is CUDA memory managed?
  4. How to measure the inner kernel time in NVIDIA CUDA?
  5. How do CUDA blocks/warps/threads map onto CUDA cores?
  6. How to verify CuDNN installation?
  7. Understanding CUDA grid dimensions, block dimensions and threads organization (simple explanation) [closed]
  8. What can I do against ‘CUDA driver version is insufficient for CUDA runtime version’?
  9. What is a bank conflict? (Doing Cuda/OpenCL programming)
  10. nvidia-smi Volatile GPU-Utilization explanation?
  11. How do I use Nvidia Multi-process Service (MPS) to run multiple non-MPI CUDA applications?
  12. How do I select which GPU to run a job on?
  13. CUDA determining threads per block, blocks per grid
  14. How does CUDA assign device IDs to GPUs?
  15. What is the canonical way to check for errors using the CUDA runtime API?
  16. How do I choose grid and block dimensions for CUDA kernels?
  17. Unspecified launch failure on Memcpy
  18. Different CUDA versions shown by nvcc and NVIDIA-smi
  19. What is the correct version of CUDA for my nvidia driver?
  20. Thrust inside user written kernels
  21. Can anyone provide sample code demonstrating the use of 16 bit floating point in cuda?
  22. CUDA: How many concurrent threads in total?
  23. CUDA apps time out & fail after several seconds – how to work around this?
  24. Why is NVIDIA Pascal GPUs slow on running CUDA Kernels when using cudaMallocManaged
  25. multi-GPU basic usage
  26. How to create a CUDA context?
  27. Why does my CUDA kernel crash (unspecified launch failure) with a different dataset size?
  28. Inconsistency of IDs between ‘nvidia-smi -L’ and cuDeviceGetName()
  29. CUDA allocation alignment is 256 bytes – seriously?
  30. Which Compute Capability is supported by which CUDA versions?

Leave a Comment