docker-compose.yml

version: '3'

services:
  grafana:
    image: grafana/grafana:8.4.4
    ports:
      - '3000:3000'
    volumes:
      - './grafana/config/grafana.ini:/etc/grafana/grafana.ini'
      - './grafana/config/dashboards.yaml:/etc/grafana/provisioning/dashboards/dashboards.yaml'
      - './grafana/dashboards/:/var/lib/grafana/dashboards/'
      - './grafana/datasources/:/etc/grafana/provisioning/datasources/'

  prometheus:
    image: prom/prometheus:v2.34.0-rc.2
    ports:
      - '9090:9090'
    volumes:
      - './prometheus/config:/etc/prometheus'

  dcgm-exporter:
    image: nvcr.io/nvidia/k8s/dcgm-exporter:2.3.5-2.6.5-ubuntu20.04
    command: dcgm-exporter -f /etc/dcgm-exporter/dcp-metrics-included.csv
    volumes:
      - './dcgm-exporter/dcp-metrics-included.csv:/etc/dcgm-exporter/dcp-metrics-included.csv'

  triton-inference-server:
    build: inference-server
    command: tritonserver --model-repository=/model_repository --model-control-mode poll --exit-on-error 0 --repository-poll-secs 3
    shm_size: '2gb'
    environment:
      - CUDA_MODULE_LOADING="LAZY"
    volumes:
      - './notebooks/code/model_repository/:/model_repository'

  notebook-app:
    command: jupyter lab -y --allow-root --no-browser --ip='*' --NotebookApp.token='' --NotebookApp.password='' --port=8884 --notebook-dir=/opt/project/
    depends_on:
      - triton-inference-server
    build: notebooks
    shm_size: '2gb'
    environment:
      - CUDA_MODULE_LOADING="LAZY"
    volumes:
      - './notebooks/code/:/opt/project'
    ports:
      - "7860:7860" # gradio UI port
      - "8884:8884" # Juypter port