rag-retrieval-benchmark/docker-compose.yml at main · RAG-Implementation/rag-retrieval-benchmark · GitHub

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
# Docker Compose stack for rag-retrieval-benchmark.
#
# Services:
#   qdrant  - vector database (CPU only)
#   jupyter - JupyterLab with GPU access, for running the phase notebooks
#   api     - FastAPI service exposing retrieval/evaluation endpoints (GPU access)
#
# The jupyter and api services share one image built from docker/Dockerfile.
# GPU access uses the Compose device-reservation syntax and requires the NVIDIA
# Container Toolkit on the host.

services:
  qdrant:
    image: qdrant/qdrant:latest
    container_name: rrb-qdrant
    ports:
      - "6333:6333" # REST API
      - "6334:6334" # gRPC API
    volumes:
      - qdrant_storage:/qdrant/storage
    healthcheck:
      # Qdrant exposes a readiness endpoint at /readyz.
      test: ["CMD", "bash", "-lc", "exec 3<>/dev/tcp/localhost/6333"]
      interval: 10s
      timeout: 5s
      retries: 10

  jupyter:
    build:
      context: .
      dockerfile: docker/Dockerfile
    image: rag-retrieval-benchmark:latest
    container_name: rrb-jupyter
    command: >
      jupyter lab
      --ip=0.0.0.0
      --port=8888
      --no-browser
      --allow-root
      --ServerApp.token=''
      --ServerApp.root_dir=/app
    ports:
      - "8888:8888"
    env_file:
      - .env
    volumes:
      - .:/app
      - hf_cache:/app/.cache/huggingface
    depends_on:
      qdrant:
        condition: service_started
    # GPU access (NVIDIA Container Toolkit required on the host).
    deploy:
      resources:
        reservations:
          devices:
            - driver: nvidia
              count: all
              capabilities: [gpu]

  api:
    build:
      context: .
      dockerfile: docker/Dockerfile
    image: rag-retrieval-benchmark:latest
    container_name: rrb-api
    command: >
      uvicorn rag_retrieval_benchmark.api.main:app
      --host 0.0.0.0
      --port 8000
    ports:
      - "8000:8000"
    env_file:
      - .env
    volumes:
      - .:/app
      - hf_cache:/app/.cache/huggingface
    depends_on:
      qdrant:
        condition: service_started
    deploy:
      resources:
        reservations:
          devices:
            - driver: nvidia
              count: all
              capabilities: [gpu]

volumes:
  qdrant_storage:
  hf_cache: