Page tree

Versions Compared

Key

  • This line was added.
  • This line was removed.
  • Formatting was changed.

...

Note

These jobs require custom code capable of leveraging GPU resources.

The GPUs available on the Rapid Access Cloud are Tesla provides 3 types of GPUs:

  • Titan K80 GPUs

...

  • with 2,496 CUDA cores and 12 GB of RAM
  • Titan XP GPUs with 3,840 CUDA cores and 12 GB of RAM
  • Titan RTX GPUs with 4608 CUDA cores and 24 GB of RAM.

Each default Rapid Access Cloud allocation has enough resources to allow a user to launch 1 VM capable of accessing 1 GPU and are available on a first come, first serve basis.

Who

The GPUs are available to any Albertan to use for pre-production, pre-commercial research, experimentation, or testing purposes. To access the default allocation of 1 GPU instance, simply launch the GPU in the Edmonton region of the Rapid Access Cloud.

...

We provide several GPU CLI only images with NVIDIA drivers and CUDA pre-installed:

Image NameCUDA Version
Ubuntu
14.048.1Ubuntu
16.0410.1
Ubuntu 18.0410.1
CentOS 710.1


How do I free up my GPU instance when it’s not in use?

...

If GPUs are not available, you will receive an error that it could not launch. However your instance's data will not be lost. Please try again later as GPUs do become available.