You can run the run command to allocate GPUs for a single application. The application runs in the entire memory resource of the GPUs.
All GPUs that are requested by using the run command must be allocated from a single vSphere Bitfusion server, and the server must list the GPUs as separate devices with different PCIe addresses.
For example, the AI application, asimov_i.py, takes two arguments: the number of GPUs and a batch size.
- When the application expects 1 GPU, run bitfusion run -n 1 -- python asimov_i.py --num_gpus=1 --batchsz=64
- When the application expects 2 GPUs, run bitfusion run -n 2 -- python asimov_i.py --num_gpus=2 --batchsz=64
By default, vSphere Bitfusion waits for 30 minutes for enough GPUs to be available. To modify the default interval, use the
--timeout value, -t value argument. Enter the timeout in seconds or time and unit, such as seconds (s), minutes (m), and hours (h).