Tensorflow 에서 GPU 가능 여부 확인 법

Tensorflow가 내 GPU를 활용하고 있는지 확인하려면, tensorflow에서 제공하는 device_lib 라이브러리를 활용하면 된다.

(mrc) [root@nipa2019-0010 mrc] python
Python 3.6.8 |Anaconda, Inc.| (default, Dec 30 2018, 01:22:34) 
[GCC 7.3.0] on linux
Type "help", "copyright", "credits" or "license" for more information.
>>> from tensorflow.python.client import device_lib
>>> device_lib.list_local_devices()

from tensorflow.python.client import device_lib

device_lib.list_local_devices()

 

위와 같은 명령어를 수행하면 필자의 경우 아래와 같은 정보를 화면에 보여주었다.

2020-06-15 14:57:48.947748: I tensorflow/core/platform/cpu_feature_guard.cc:141] Your CPU supports instructions that this TensorFlow binary was not compiled to use: SSE4.1 SSE4.2 AVX AVX2 FMA
2020-06-15 14:57:49.156328: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:964] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero
2020-06-15 14:57:49.157817: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1411] Found device 0 with properties: 
name: Tesla V100-PCIE-32GB major: 7 minor: 0 memoryClockRate(GHz): 1.38
pciBusID: 0000:00:05.0
totalMemory: 31.75GiB freeMemory: 31.44GiB
2020-06-15 14:57:49.308937: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:964] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero
2020-06-15 14:57:49.310258: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1411] Found device 1 with properties: 
name: Tesla V100-PCIE-32GB major: 7 minor: 0 memoryClockRate(GHz): 1.38
pciBusID: 0000:00:06.0
totalMemory: 31.75GiB freeMemory: 31.44GiB
2020-06-15 14:57:49.310338: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1490] Adding visible gpu devices: 0, 1
2020-06-15 14:57:50.139037: I tensorflow/core/common_runtime/gpu/gpu_device.cc:971] Device interconnect StreamExecutor with strength 1 edge matrix:
2020-06-15 14:57:50.139088: I tensorflow/core/common_runtime/gpu/gpu_device.cc:977]      0 1 
2020-06-15 14:57:50.139137: I tensorflow/core/common_runtime/gpu/gpu_device.cc:990] 0:   N N 
2020-06-15 14:57:50.139148: I tensorflow/core/common_runtime/gpu/gpu_device.cc:990] 1:   N N 
2020-06-15 14:57:50.139312: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1103] Created TensorFlow device (/device:GPU:0 with 30503 MB memory) -> physical GPU (device: 0, name: Tesla V100-PCIE-32GB, pci bus id: 0000:00:05.0, compute capability: 7.0)
2020-06-15 14:57:50.586570: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1103] Created TensorFlow device (/device:GPU:1 with 30503 MB memory) -> physical GPU (device: 1, name: Tesla V100-PCIE-32GB, pci bus id: 0000:00:06.0, compute capability: 7.0)
[name: "/device:CPU:0"
device_type: "CPU"
memory_limit: 268435456
locality {
}
incarnation: 869474478715071609
, name: "/device:GPU:0"
device_type: "GPU"
memory_limit: 31985477223
locality {
  bus_id: 1
  links {
  }
}
incarnation: 12179460544041194412
physical_device_desc: "device: 0, name: Tesla V100-PCIE-32GB, pci bus id: 0000:00:05.0, compute capability: 7.0"
, name: "/device:GPU:1"
device_type: "GPU"
memory_limit: 31985477223
locality {
  bus_id: 1
  links {
  }
}
incarnation: 3329334536086686166
physical_device_desc: "device: 1, name: Tesla V100-PCIE-32GB, pci bus id: 0000:00:06.0, compute capability: 7.0"
]

GPU가 잡힌 것을 확인 할 수 있고, 이는 이 GPU를 이용하여 작업이 가능하다는 의미이다.