timeplus-gpu-mointoring-stack

docker-compose.yaml

services:
  # Ollama service with GPU support
  ollama:
    container_name: ollama
    image: ollama/ollama:latest
    ports:
      - "11434:11434"
    volumes:
      - ollama-data:/root/.ollama
    environment:
      - OLLAMA_KEEP_ALIVE=24h
    deploy:
      resources:
        reservations:
          devices:
            - driver: nvidia
              count: all  # Use all available GPUs
              capabilities: [gpu]
    restart: unless-stopped
    
  openwebui:
    container_name: openwebui
    image: ghcr.io/open-webui/open-webui:main
    ports:
      - "3000:8080"
    environment:
      - OLLAMA_API_BASE_URL=http://ollama:11434
    volumes:
      - openwebui-data:/app/backend/data
    depends_on:
      - ollama
    restart: unless-stopped
  
  redpanda:
    image: redpandadata/redpanda:v23.3.4
    container_name: redpanda
    command:
      - redpanda start
      - --smp 1
      - --memory 1G
      - --reserve-memory 0M
      - --overprovisioned
      - --node-id 0
      - --check=false
      - --kafka-addr PLAINTEXT://0.0.0.0:9092,OUTSIDE://0.0.0.0:19092
      - --advertise-kafka-addr PLAINTEXT://redpanda:9092,OUTSIDE://localhost:19092
      - --schema-registry-addr 0.0.0.0:8081
      - --pandaproxy-addr 0.0.0.0:8082
      - --advertise-pandaproxy-addr localhost:8082
    ports:
      - "19092:19092"
      - "9644:9644"  # Admin API
      - "8081:8081"  # Schema Registry
    volumes:
      - redpanda-data:/var/lib/redpanda
    environment:
      KAFKA_AUTO_CREATE_TOPICS_ENABLE: "true"

  dcgm-exporter:
    image: nvcr.io/nvidia/k8s/dcgm-exporter:latest
    runtime: nvidia
    devices:
      - /dev/nvidiactl:/dev/nvidiactl
      - /dev/nvidia0:/dev/nvidia0  # Add more for multiple GPUs
    ports:
      - "9400:9400"
    restart: always

  # Vector to collect and forward metrics
  vector:
    image: timberio/vector:latest-alpine
    volumes:
      - ./vector.toml:/etc/vector/vector.toml:ro
    environment:
      - VECTOR_CONFIG=/etc/vector/vector.toml
    ports:
      - "8686:8686"  # Vector API server
    depends_on:
      - dcgm-exporter
    restart: always

  timeplus:
    image: timeplus/timeplus-enterprise:2.7.7
    ports:
      - 8000:8000
    volumes:
      - timeplus-data:/timeplus/data/

volumes:
  ollama-data:
  openwebui-data:
  redpanda-data:
  timeplus-data: