BitNet-rs/docker-compose.yml at main · EffortlessMetrics/BitNet-rs · 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
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
# Docker Compose for BitNet-rs
#
# Usage:
#   docker compose up bitnet-cpu            # CPU inference server
#   docker compose --profile gpu up         # GPU inference server
#   docker compose --profile production up  # CPU + nginx reverse proxy
#   docker compose --profile monitoring up  # CPU + Prometheus + Grafana
#
# Environment variables (set in shell or .env file):
#   GIT_SHA        - Git commit hash for OCI labels
#   GIT_BRANCH     - Git branch name for OCI labels
#   GIT_DESCRIBE   - Git describe output for version label
#   BITNET_THREADS - Rayon thread count (default: 4)
#   CUDA_DEVICE    - GPU device index (default: 0)
# ─────────────────────────────────────────────────────────────────

services:
  # ── CPU inference server ──────────────────────────────────────────
  bitnet-cpu:
    build:
      context: .
      dockerfile: Dockerfile
      target: runtime
      args:
        VCS_REF: ${GIT_SHA:-unknown}
        VCS_BRANCH: ${GIT_BRANCH:-unknown}
        VCS_DESCRIBE: ${GIT_DESCRIBE:-unknown}
    image: bitnet-rs:cpu
    container_name: bitnet-cpu
    ports:
      - "8080:8080"
    volumes:
      - ./models:/models:ro        # Model files (read-only)
      - bitnet_data:/data           # Writable data (receipts, logs)
    environment:
      - RUST_LOG=info               # Log level: error|warn|info|debug|trace
      - BITNET_THREADS=${BITNET_THREADS:-4}
    command: ["server"]
    restart: unless-stopped
    user: "10001:10001"
    cap_drop:
      - ALL
    security_opt:
      - no-new-privileges:true
    read_only: true
    tmpfs:
      - /tmp:rw,noexec,nosuid,nodev
    healthcheck:
      test: ["CMD", "curl", "-sf", "http://localhost:8080/health"]
      interval: 30s
      timeout: 10s
      retries: 3
      start_period: 30s
    deploy:
      resources:
        limits:
          cpus: '4'
          memory: 8G
        reservations:
          cpus: '2'
          memory: 4G

  # ── GPU (CUDA) inference server ───────────────────────────────────
  bitnet-gpu:
    build:
      context: .
      dockerfile: Dockerfile
      target: runtime-gpu
      args:
        VCS_REF: ${GIT_SHA:-unknown}
        VCS_BRANCH: ${GIT_BRANCH:-unknown}
        VCS_DESCRIBE: ${GIT_DESCRIBE:-unknown}
    image: bitnet-rs:gpu
    container_name: bitnet-gpu
    ports:
      - "8081:8080"
    volumes:
      - ./models:/models:ro
      - bitnet_data_gpu:/data
    environment:
      - RUST_LOG=info
      - CUDA_VISIBLE_DEVICES=${CUDA_DEVICE:-0}
    command: ["server"]
    restart: unless-stopped
    user: "10001:10001"
    cap_drop:
      - ALL
    security_opt:
      - no-new-privileges:true
    read_only: true
    tmpfs:
      - /tmp:rw,noexec,nosuid,nodev
    healthcheck:
      test: ["CMD", "curl", "-sf", "http://localhost:8080/health"]
      interval: 30s
      timeout: 10s
      retries: 3
      start_period: 30s
    deploy:
      resources:
        limits:
          cpus: '4'
          memory: 16G
        reservations:
          devices:
            - driver: nvidia
              count: 1
              capabilities: [gpu]
    profiles:
      - gpu

  # ── Reverse proxy (optional) ─────────────────────────────────────
  nginx:
    image: nginx:1.25-alpine
    container_name: bitnet-lb
    ports:
      - "80:80"
      - "443:443"
    volumes:
      - ./nginx.conf:/etc/nginx/nginx.conf:ro
      - ./certs:/etc/nginx/certs:ro
    depends_on:
      - bitnet-cpu
    restart: unless-stopped
    profiles:
      - production

  # ── Monitoring (optional) ─────────────────────────────────────────
  prometheus:
    image: prom/prometheus:v2.53.0
    container_name: bitnet-prometheus
    ports:
      - "9090:9090"
    volumes:
      - ./prometheus.yml:/etc/prometheus/prometheus.yml:ro
      - prometheus_data:/prometheus
    command:
      - '--config.file=/etc/prometheus/prometheus.yml'
      - '--storage.tsdb.path=/prometheus'
    restart: unless-stopped
    profiles:
      - monitoring

  grafana:
    image: grafana/grafana:10.4.3
    container_name: bitnet-grafana
    ports:
      - "3000:3000"
    volumes:
      - grafana_data:/var/lib/grafana
      - ./grafana/dashboards:/etc/grafana/provisioning/dashboards:ro
      - ./grafana/datasources:/etc/grafana/provisioning/datasources:ro
    environment:
      - GF_SECURITY_ADMIN_PASSWORD=admin
      - GF_USERS_ALLOW_SIGN_UP=false
    restart: unless-stopped
    profiles:
      - monitoring

volumes:
  bitnet_data:
  bitnet_data_gpu:
  prometheus_data:
  grafana_data:

networks:
  default:
    name: bitnet-network
    driver: bridge