agg_8b.yaml 2.56 KB
Newer Older
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
# SPDX-FileCopyrightText: Copyright (c) 2025 NVIDIA CORPORATION & AFFILIATES. All rights reserved.
# SPDX-License-Identifier: Apache-2.0

apiVersion: nvidia.com/v1alpha1
kind: DynamoGraphDeployment
metadata:
  name: vllm-agg
spec:
  services:
    Frontend:
      dynamoNamespace: vllm-agg
      componentType: main
      replicas: 1
      livenessProbe:
        httpGet:
          path: /health
          port: 8000
        initialDelaySeconds: 20
        periodSeconds: 5
        timeoutSeconds: 5
        failureThreshold: 3
      readinessProbe:
        exec:
          command:
            - /bin/sh
            - -c
            - 'curl -s http://localhost:8000/health | jq -e ".status == \"healthy\""'
        initialDelaySeconds: 60
        periodSeconds: 60
        timeoutSeconds: 30
        failureThreshold: 10
      resources:
        requests:
          cpu: "16"
          memory: "10Gi"
        limits:
          cpu: "128"
          memory: "100Gi"
      extraPodSpec:
        mainContainer:
41
          image: my-registry/vllm-runtime:my-tag
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
          workingDir: /workspace/components/backends/vllm
          command:
            - /bin/sh
            - -c
          args:
            - "python3 -m dynamo.frontend --http-port 8000"
    VllmDecodeWorker:
      dynamoNamespace: vllm-agg
      envFromSecret: hf-token-secret
      componentType: worker
      replicas: 1
      livenessProbe:
        httpGet:
          path: /live
          port: 9090
        periodSeconds: 5
        timeoutSeconds: 30
        failureThreshold: 1
      readinessProbe:
        httpGet:
          path: /health
          port: 9090
        periodSeconds: 10
        timeoutSeconds: 30
        failureThreshold: 60
      resources:
        requests:
          cpu: "16"
          memory: "20Gi"
          gpu: "1"
        limits:
          cpu: "128"
          memory: "100Gi"
          gpu: "1"
      envs:
        - name: DYN_SYSTEM_ENABLED
          value: "true"
        - name: DYN_SYSTEM_PORT
          value: "9090"
81
82
        - name: DYN_SYSTEM_USE_ENDPOINT_HEALTH_STATUS
          value: "[\"generate\"]"
83
84
85
86
87
88
89
90
      extraPodSpec:
        mainContainer:
          startupProbe:
            httpGet:
              path: /health
              port: 9090
            periodSeconds: 10
            failureThreshold: 60
91
          image: my-registry/vllm-runtime:my-tag
92
93
94
95
96
97
          workingDir: /workspace/components/backends/vllm
          command:
            - /bin/sh
            - -c
          args:
            - "python3 -m dynamo.vllm --model nvidia/Llama-3.1-8B-Instruct-FP8 --no-enable-prefix-caching 2>&1 | tee /tmp/vllm.log"