oneflow serving一定需要gpu的设备才能部署吗？

[aurora5162]

Description

在本机上（ubuntu20.0.4，12核cpu），执行如下命令： sudo docker run --rm -p 8001:8001 -p8000:8000 -p 8002:8002 -v /home/openmldb/jd-recommendation/oneflow_process/model:/models -v /home/openmldb/jd-recommendation/oneflow_process/persistent:/root/demo/persistent oneflowinc/oneflow-serving:nightly bash -c '/opt/tritonserver/bin/tritonserver --model-repository=/models' 报错如下： Error: Failed to initialize NVML W0929 05:00:15.927491 1 metrics.cc:221] DCGM unable to start: DCGM initialization error W0929 05:00:15.927774 1 pinned_memory_manager.cc:236] Unable to allocate pinned system memory, pinned memory pool will not be available: CUDA driver version is insufficient for CUDA runtime version I0929 05:00:15.927814 1 cuda_memory_manager.cc:115] CUDA memory pool disabled E0929 05:00:15.929605 1 model_repository_manager.cc:1890] Poll failed for model directory 'embedding': instance group embedding_0 of model embedding specifies invalid or unsupported gpu id 0. GPUs with at least the minimum required CUDA compute compatibility of 6.000000 are: I0929 05:00:15.929647 1 server.cc:522] I0929 05:00:15.929837 1 server.cc:252] Waiting for in-flight requests to complete. I0929 05:00:15.929847 1 server.cc:267] Timeout 30: Found 0 live models and 0 in-flight non-inference requests error: creating server: Internal - failed to load all models 麻烦问下作者 ，一定需要gpu的设备才能部署oneflow serving 吗？

Alternatives

No response

[jackalcooper]

是的

[aurora5162]

是的

好的，了解了，谢谢