Inquiry: Configuring `xinference` in Docker to Utilize Multiple CPU Cores

[stoplyy]

Describe the inquiry

I need guidance on how to configure the xinference process within a Docker container so that it can utilize multiple CPU cores. Currently, although I have 64 CPU cores available, the process seems to be using only one core. My launch command is:

docker run -e XINFERENCE_MODEL_SRC=modelscope -p 9998:9997 -v /root/lyy/models:/app/models xprobe/xinference xinference-local -H 0.0.0.0 --log-level debug

我希望能够充分利用我的系统资源，以提高处理效率。

I am looking to fully utilize my system resources to enhance processing efficiency.

Details Required for Assistance

To help provide specific guidance, please provide the following details:

1. Operating System: Ubuntu
2. Docker Image Used: xprobe/xinference:latest-cpu
3. Current CPU Configuration: 64 cores, 456GB RAM
4. Detailed description of the current launch command and configuration.
5. Changes in configuration already tried and methods attempted to increase CPU usage.

Expected Configuration

I would like the xinference process to utilize more CPU cores (e.g., at least 32 cores) in order to improve task processing speed and overall application performance.

Additional context

This application is a critical part of data processing, where processing efficiency directly affects operational efficiency. The current limitation to a single core significantly reduces its performance, so any advice on increasing CPU utilization would be extremely valuable.