16:00:27.311963: I tensorflow/core/common_runtime/gpu/gpu_:1241] Created TensorFlow device (/job:localhost/replica:0/task:0/device:GPU:0 with 6706 MB memory) -> physical GPU (device: 0, name: NVIDIA GeForce RTX 2080, pci bus id: 0000:65:00.0, compute capability: 7.5) Your kernel may not have been built with NUMA support. 16:00:27.311180: I tensorflow/core/common_runtime/gpu/gpu_:1324] Could not identify NUMA node of platform GPU id 0, defaulting to 0. 16:00:27.309024: I tensorflow/core/common_runtime/gpu/gpu_:1096] Device interconnect StreamExecutor with strength 1 edge matrix: 16:00:26.149385: I tensorflow/stream_executor/platform/default/dso_:44] Successfully opened dynamic library libcudart.so.10.1 Your kernel may have been built without NUMA support. 16:00:25.573058: I tensorflow/stream_executor/platform/default/dso_:44] Successfully opened dynamic library libcuda.so.1 16:00:14.357695: I tensorflow/stream_executor/platform/default/dso_:44] Successfully opened dynamic library libnvinfer_plugin.so.6 16:00:14.343853: I tensorflow/stream_executor/platform/default/dso_:44] Successfully opened dynamic library libnvinfer.so.6 Notebook tensorflow-tutorials/classification.ipynb is not trusted Writing notebook-signing key to /root/.local/share/jupyter/notebook_secret Within the Jupyter notebook WSL 2 container to see the work accelerated by the GPU of
Navigate to the Cell menu and select the Run All item, then check the log Choose any of the tutorials for this example. Ensure that you replaceĬonnecting to the Jupyter notebook from the browser. Start developing with the Jupyter notebook. To access the notebook, open this file in a browser:įile:///root/.local/share/jupyter/runtime/nbserver-1-open.htmlĪfter the URL is available from the console output, input the URL into your browser to Use Control-C to stop this server and shut down all kernels (twice to skip confirmation). Serving notebooks from local directory: /tf Jupyter_http_over_ws extension initialized. Writing notebook server cookie secret to /root/.local/share/jupyter/runtime/notebook_cookie_secret To avoid this, run the container by specifying your user's userid:
Mounted volumes to be created as the root user on your host machine. WARNING: You are running this container as root, which can cause new files in = 6483.371 single-precision GFLOP/s at 20 flops per interaction = 324.169 billion interactions per second > Compute 7.5 CUDA device: Ĥ7104 bodies, total time for 10 iterations: 68.445 ms GPU Device 0: "Turing" with compute capability 7.5 > Single precision floating point simulation Results may vary when GPU Boost is enabled.
NOTE: The CUDA Samples are not meant for performance measurements. tipsy= (load a tipsy model file for simulation) compare (compares simulation results running once on the default GPU and once on the CPU) numdevices= (where i=(number of CUDA devices > 0) to use for simulation) numbodies= (number of bodies (>= 1) to run in simulation) benchmark (run benchmark to measure performance) hostmem (stores simulation data in host memory) fp64 (use double precision floating point values for simulation) fullscreen (run n-body simulation in fullscreen mode) Run "nbody -benchmark " to measure performance. $ docker run -gpus all nvcr.io/nvidia/k8s/cuda-sample:nbody nbody -gpu -benchmark
0 kommentar(er)
