-
Notifications
You must be signed in to change notification settings - Fork 0
/
docker-compose.yml
118 lines (110 loc) · 2.37 KB
/
docker-compose.yml
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
version: '2'
services:
gradio_frontend:
image: pytorch/pytorch:2.4.1-cuda12.4-cudnn9-runtime
container_name: gradio_frontend
build:
context: ./gradio
ports:
- "7860:7860"
networks:
- ai_network
volumes:
- ./gradio:/app
environment:
- ROLE=frontend
depends_on:
- master
- redis
cpu_shares: 1024
mem_limit: 4g
master:
image: pytorch/pytorch:2.4.1-cuda12.4-cudnn9-runtime
container_name: inference_master
build:
context: ./master
ports:
- "5000:5000"
networks:
- ai_network
volumes:
- ./master:/app
environment:
- ROLE=master
depends_on:
- slave1
- slave2
- slave3
- slave4
cpu_shares: 1024
mem_limit: 4g
slave1:
image: pytorch/pytorch:2.4.1-cuda12.4-cudnn9-runtime
container_name: inference_slave1
build:
context: ./slave
networks:
- ai_network
volumes:
- ./slave:/app
environment:
- CUDA_VISIBLE_DEVICES=0
- ROLE=slave
- WORKER_NAME=slave1
cpu_shares: 512
mem_limit: 2g
slave2:
image: pytorch/pytorch:2.4.1-cuda12.4-cudnn9-runtime
container_name: inference_slave2
build:
context: ./slave
networks:
- ai_network
volumes:
- ./slave:/app
environment:
- CUDA_VISIBLE_DEVICES=1
- ROLE=slave
- WORKER_NAME=slave2
cpu_shares: 512
mem_limit: 2g
slave3:
image: pytorch/pytorch:2.4.1-cuda12.4-cudnn9-runtime
container_name: inference_slave3
build:
context: ./slave
networks:
- ai_network
volumes:
- ./slave:/app
environment:
- CUDA_VISIBLE_DEVICES=2
- ROLE=slave
- WORKER_NAME=slave3
cpu_shares: 512
mem_limit: 2g
slave4:
image: pytorch/pytorch:2.4.1-cuda12.4-cudnn9-runtime
container_name: inference_slave4
build:
context: ./slave
networks:
- ai_network
volumes:
- ./slave:/app
environment:
- CUDA_VISIBLE_DEVICES=3
- ROLE=slave
- WORKER_NAME=slave4
cpu_shares: 512
mem_limit: 2g
redis:
image: redis:7.4.1-bookworm
container_name: redis
networks:
- ai_network
ports:
- "6379:6379"
networks:
ai_network:
driver: bridge