我已经安装了*tensorflow-model-server 1.15版的apt库。我所有的 tensorflow 程序似乎都在 GPU 上运行，但不知何故 tensorflow-model-server 没有。之前我曾经在我的日志文件中获得以下内容，但在我的新实例中没有获得任何与 gpu 或 cuda 相关的日志。
> 2021-04-09 05:15:32.044033: I external/org_tensorflow/tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcuda.so.1
> 2021-04-09 05:15:32.115742: I external/org_tensorflow/tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:983] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero
> 2021-04-09 05:15:32.116637: I external/org_tensorflow/tensorflow/core/common_runtime/gpu/gpu_device.cc:1639] Found device 0 with properties: name: Tesla T4 major: 7 minor: 5 memoryClockRate(GHz): 1.59 pciBusID: 0000:00:1e.0
> 2021-04-09 05:15:32.116673: I external/org_tensorflow/tensorflow/stream_executor/platform/default/dlopen_checker_stub.cc:25] GPU libraries are statically linked, skip dlopen check.
> 2021-04-09 05:15:32.116739: I external/org_tensorflow/tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:983] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero
> 2021-04-09 05:15:32.117649: I external/org_tensorflow/tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:983] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero
> 2021-04-09 05:15:32.118491: I external/org_tensorflow/tensorflow/core/common_runtime/gpu/gpu_device.cc:1767] Adding visible gpu devices: 0
Tensorflow-gpu = 1.15.2
Cuda = 10.0
库德恩 = 7.5.1.10