How to check if tensorflow is using all available GPU's

Question

I am learning to use Tensorflow for object detection. To speed up the training process, I have taken a AWS g3.16xlarge instance which has 4 GPUs. I am using the following code to run training process:

export CUDA_VISIBLE_DEVICES=0,1,2,3
 python object_detection/train.py --logtostderr --pipeline_config_path=/home/ubuntu/builder/rcnn.config --train_dir=/home/ubuntu/builder/experiments/training/

Inside the rcnn.config - i have set the batch-size = 1. During runtime I get the following output:

console output

2018-11-09 07:25:50.104310: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1045] Device peer to peer matrix
2018-11-09 07:25:50.104385: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1051] DMA: 0 1 2 3 
2018-11-09 07:25:50.104395: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1061] 0:   Y N N N 
2018-11-09 07:25:50.104402: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1061] 1:   N Y N N 
2018-11-09 07:25:50.104409: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1061] 2:   N N Y N 
2018-11-09 07:25:50.104416: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1061] 3:   N N N Y 
2018-11-09 07:25:50.104429: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1120] Creating TensorFlow device (/device:GPU:0) -> (device: 0, name: Tesla M60, pci bus id: 0000:00:1b.0, compute capability: 5.2)
2018-11-09 07:25:50.104439: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1120] Creating TensorFlow device (/device:GPU:1) -> (device: 1, name: Tesla M60, pci bus id: 0000:00:1c.0, compute capability: 5.2)
2018-11-09 07:25:50.104446: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1120] Creating TensorFlow device (/device:GPU:2) -> (device: 2, name: Tesla M60, pci bus id: 0000:00:1d.0, compute capability: 5.2)
2018-11-09 07:25:50.104455: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1120] Creating TensorFlow device (/device:GPU:3) -> (device: 3, name: Tesla M60, pci bus id: 0000:00:1e.0, compute capability: 5.2)

When I run nvidia-smi, I get the following output: nvidia-smi output

+-----------------------------------------------------------------------------+
| NVIDIA-SMI 375.26                 Driver Version: 375.26                    |
|-------------------------------+----------------------+----------------------+
| GPU  Name        Persistence-M| Bus-Id        Disp.A | Volatile Uncorr. ECC |
| Fan  Temp  Perf  Pwr:Usage/Cap|         Memory-Usage | GPU-Util  Compute M. |
|===============================+======================+======================|
|   0  Tesla M60           Off  | 0000:00:1B.0     Off |                    0 |
| N/A   52C    P0   129W / 150W |   7382MiB /  7612MiB |     92%      Default |
+-------------------------------+----------------------+----------------------+
|   1  Tesla M60           Off  | 0000:00:1C.0     Off |                    0 |
| N/A   33C    P0    38W / 150W |   7237MiB /  7612MiB |      0%      Default |
+-------------------------------+----------------------+----------------------+
|   2  Tesla M60           Off  | 0000:00:1D.0     Off |                    0 |
| N/A   40C    P0    38W / 150W |   7237MiB /  7612MiB |      0%      Default |
+-------------------------------+----------------------+----------------------+
|   3  Tesla M60           Off  | 0000:00:1E.0     Off |                    0 |
| N/A   34C    P0    39W / 150W |   7237MiB /  7612MiB |      0%      Default |
+-------------------------------+----------------------+----------------------+

+-----------------------------------------------------------------------------+
| Processes:                                                       GPU Memory |
|  GPU       PID  Type  Process name                               Usage      |
|=============================================================================|
|    0     97860    C   python                                        7378MiB |
|    1     97860    C   python                                        7233MiB |
|    2     97860    C   python                                        7233MiB |
|    3     97860    C   python                                        7233MiB |
+-----------------------------------------------------------------------------+

and **nvidia-smi dmon** provides the following output:

# gpu   pwr  temp    sm   mem   enc   dec  mclk  pclk
# Idx     W     C     %     %     %     %   MHz   MHz
    0   158    69    90    69     0     0  2505  1177
    1    38    36     0     0     0     0  2505   556
    2    38    45     0     0     0     0  2505   556
    3    39    37     0     0     0     0  2505   556

I am confused with each of the output. While I read the console output as the program is recognizing the availability of 4 different gpus, in the nvidia-smi output the volatile GPU-Util percentage is shown only for the first GPU and for the rest it is zero. However the same table prints memory usage for all the 4 gpu's at the bottom. And the nvidia-smi dmon prints the sm values only for first gpu and for the others it is zero. From this blog I understand the zero in dmon indicates that GPU is free.

What I want to understand is, does the train.py utilizes all the 4 GPU's that I have in my instance. If it is not utilizing all the GPU's how do I ensure the object_detection/train.py of tensorflow is optimized for all the GPU's.

Answer 1

Python code to check if GPU is found and available for using with tensorflow:

## Libraries import
import tensorflow as tf

## Test GPU
device_name = tf.test.gpu_device_name()
if device_name != '/device:GPU:0':
  raise SystemError('GPU device not found')
print('Found GPU at: {}'.format(device_name))
print('')
config = tf.ConfigProto()
config.gpu_options.allow_growth = True

Answer 2

Check if it's returning list of all GPUs.

tf.test.gpu_device_name()

Returns the name of a GPU device if available or the empty string.

then you can do something like this to use all the available GPUs.

# Creates a graph.
c = []
for d in ['/device:GPU:2', '/device:GPU:3']:
  with tf.device(d):
    a = tf.constant([1.0, 2.0, 3.0, 4.0, 5.0, 6.0], shape=[2, 3])
    b = tf.constant([1.0, 2.0, 3.0, 4.0, 5.0, 6.0], shape=[3, 2])
    c.append(tf.matmul(a, b))
with tf.device('/cpu:0'):
  sum = tf.add_n(c)
# Creates a session with log_device_placement set to True.
sess = tf.Session(config=tf.ConfigProto(log_device_placement=True))
# Runs the op.
print(sess.run(sum))

You see below output:

Device mapping:
/job:localhost/replica:0/task:0/device:GPU:0 -> device: 0, name: Tesla K20m, pci bus
id: 0000:02:00.0
/job:localhost/replica:0/task:0/device:GPU:1 -> device: 1, name: Tesla K20m, pci bus
id: 0000:03:00.0
/job:localhost/replica:0/task:0/device:GPU:2 -> device: 2, name: Tesla K20m, pci bus
id: 0000:83:00.0
/job:localhost/replica:0/task:0/device:GPU:3 -> device: 3, name: Tesla K20m, pci bus
id: 0000:84:00.0
Const_3: /job:localhost/replica:0/task:0/device:GPU:3
Const_2: /job:localhost/replica:0/task:0/device:GPU:3
MatMul_1: /job:localhost/replica:0/task:0/device:GPU:3
Const_1: /job:localhost/replica:0/task:0/device:GPU:2
Const: /job:localhost/replica:0/task:0/device:GPU:2
MatMul: /job:localhost/replica:0/task:0/device:GPU:2
AddN: /job:localhost/replica:0/task:0/cpu:0
[[  44.   56.]
 [  98.  128.]]