Targeting Specific GPUs and Accessing Specific Devices

3rd chunk of `content/manuals/compose/how-tos/gpu-support.md`

a202ed2e36333fc0c6fde49ab0d7119ca903e06b864dc4750000000100000cc3

test_1  | | Processes:                                                                  |
test_1  | |  GPU   GI   CI        PID   Type   Process name                  GPU Memory |
test_1  | |        ID   ID                                                   Usage      |
test_1  | |=============================================================================|
test_1  | |  No running processes found                                                 |
test_1  | +-----------------------------------------------------------------------------+
gpu_test_1 exited with code 0

```

On machines hosting multiple GPUs, the `device_ids` field can be set to target specific GPU devices and `count` can be used to limit the number of GPU devices assigned to a service container. 

You can use `count` or `device_ids` in each of your service definitions. An error is returned if you try to combine both, specify an invalid device ID, or use a value of count that’s higher than the number of GPUs in your system.

```console
$ nvidia-smi   
+-----------------------------------------------------------------------------+
| NVIDIA-SMI 450.80.02    Driver Version: 450.80.02    CUDA Version: 11.0     |
|-------------------------------+----------------------+----------------------+
| GPU  Name        Persistence-M| Bus-Id        Disp.A | Volatile Uncorr. ECC |
| Fan  Temp  Perf  Pwr:Usage/Cap|         Memory-Usage | GPU-Util  Compute M. |
|                               |                      |               MIG M. |
|===============================+======================+======================|
|   0  Tesla T4            On   | 00000000:00:1B.0 Off |                    0 |
| N/A   72C    P8    12W /  70W |      0MiB / 15109MiB |      0%      Default |
|                               |                      |                  N/A |
+-------------------------------+----------------------+----------------------+
|   1  Tesla T4            On   | 00000000:00:1C.0 Off |                    0 |
| N/A   67C    P8    11W /  70W |      0MiB / 15109MiB |      0%      Default |
|                               |                      |                  N/A |
+-------------------------------+----------------------+----------------------+
|   2  Tesla T4            On   | 00000000:00:1D.0 Off |                    0 |
| N/A   74C    P8    12W /  70W |      0MiB / 15109MiB |      0%      Default |
|                               |                      |                  N/A |
+-------------------------------+----------------------+----------------------+
|   3  Tesla T4            On   | 00000000:00:1E.0 Off |                    0 |
| N/A   62C    P8    11W /  70W |      0MiB / 15109MiB |      0%      Default |
|                               |                      |                  N/A |
+-------------------------------+----------------------+----------------------+
```

## Access specific devices

To allow access only to GPU-0 and GPU-3 devices:

```yaml
services:
  test:
    image: tensorflow/tensorflow:latest-gpu
    command: python -c "import tensorflow as tf;tf.test.gpu_device_name()"
    deploy:
      resources:
        reservations:
          devices:
          - driver: nvidia
            device_ids: ['0', '3']
            capabilities: [gpu]

```

Title: Targeting Specific GPUs and Accessing Specific Devices

Summary

This section explains how to target specific GPUs on machines with multiple GPUs using the `device_ids` field in the Docker Compose file. It reiterates that `count` can also be used to limit the number of GPUs assigned, but `count` and `device_ids` cannot be combined. It then shows an example `nvidia-smi` output for a machine with multiple Tesla T4 GPUs. Finally, it provides a YAML configuration example demonstrating how to allow access only to GPU-0 and GPU-3 devices using the `device_ids` field within the `deploy.resources.reservations.devices` section.