disagg_planner.yaml

# 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: sglang-disagg-planner
  annotations:
    nvidia.com/enable-grove: "false"
spec:
  envs:
    - name: DYNAMO_SERVICE_CONFIG
      value: '{"Prometheus":{"global":{"scrape_interval":"5s"},"scrape_configs":[{"job_name":"prometheus","static_configs":[{"targets":["localhost:9090"]}]},{"job_name":"frontend","static_configs":[{"targets":["sglang-disagg-planner-frontend:8000"]}]}]}}'
    - name: DYNAMO_NAMESPACE
      value: "dynamo"
  services:
    Frontend:
      dynamoNamespace: dynamo
      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
      componentType: main
      replicas: 1
      resources:
        requests:
          cpu: "10"
          memory: "10Gi"
        limits:
          cpu: "32"
          memory: "40Gi"
      extraPodSpec:
        mainContainer:
          image: nvcr.io/nvidian/nim-llm-dev/sglang-runtime:hzhou-0811-1
          workingDir: /workspace/components/backends/sglang
          command: ["sh", "-c"]
          args:
            - "python3 -m dynamo.sglang.utils.clear_namespace --namespace sglang-disagg && python3 -m dynamo.frontend --http-port=8000"
    Planner:
      dynamoNamespace: dynamo
      envFromSecret: hf-token-secret
      componentType: planner
      replicas: 1
      livenessProbe:
        exec:
          command:
            - /bin/sh
            - -c
            - "exit 0"
        periodSeconds: 60
        timeoutSeconds: 30
        failureThreshold: 10
      readinessProbe:
        exec:
          command:
            - /bin/sh
            - -c
            - "exit 0"
        initialDelaySeconds: 60
        periodSeconds: 60
        timeoutSeconds: 30
        failureThreshold: 10
      resources:
        requests:
          cpu: "2"
          memory: "2Gi"
        limits:
          cpu: "8"
          memory: "16Gi"
      pvc:
        create: false
        name: profiling-pvc # Must be pre-created before deployment and SLA profiler must have been run
        mountPoint: /workspace/profiling_results
      extraPodSpec:
        mainContainer:
          image: nvcr.io/nvidian/nim-llm-dev/sglang-runtime:hzhou-0811-1
          workingDir: /workspace/components/planner/src/dynamo/planner
          args:
            - python
            - -m
            - planner_sla
            - --environment=kubernetes
            - --backend=sglang
            - --adjustment-interval=60
            - --profile-results-dir=/workspace/profiling_results
    Prometheus:
      dynamoNamespace: dynamo
      componentType: main
      replicas: 1
      envs:
        - name: PYTHONPATH
          value: "/workspace/components/planner/src"
      livenessProbe:
        exec:
          command:
            - /bin/sh
            - -c
            - "exit 0"
        periodSeconds: 60
        timeoutSeconds: 30
        failureThreshold: 10
      readinessProbe:
        exec:
          command:
            - /bin/sh
            - -c
            - "exit 0"
        initialDelaySeconds: 30
        periodSeconds: 60
        timeoutSeconds: 30
        failureThreshold: 10
      resources:
        requests:
          cpu: "2"
          memory: "2Gi"
        limits:
          cpu: "8"
          memory: "16Gi"
      extraPodSpec:
        mainContainer:
          image: nvcr.io/nvidian/nim-llm-dev/sglang-runtime:hzhou-0811-1
          workingDir: /workspace/components/backends/sglang
          command:
            - /bin/sh
            - -c
          args:
            - "python3 -m dynamo.planner.prometheus"
    SGLangDecodeWorker:
      dynamoNamespace: dynamo
      envFromSecret: hf-token-secret
      livenessProbe:
        httpGet:
          path: /live
          port: 9090
        periodSeconds: 5
        timeoutSeconds: 30
        failureThreshold: 1
      readinessProbe:
        httpGet:
          path: /health
          port: 9090
        periodSeconds: 10
        timeoutSeconds: 30
        failureThreshold: 60
      componentType: worker
      replicas: 2
      resources:
        requests:
          cpu: "10"
          memory: "20Gi"
          gpu: "1"
        limits:
          cpu: "32"
          memory: "80Gi"
          gpu: "1"
      envs:
        - name: DYN_SYSTEM_ENABLED
          value: "true"
        - name: DYN_SYSTEM_USE_ENDPOINT_HEALTH_STATUS
          value: "[\"generate\"]"
        - name: DYN_SYSTEM_PORT
          value: "9090"
      extraPodSpec:
        mainContainer:
          startupProbe:
            httpGet:
              path: /live
              port: 9090
            periodSeconds: 10
            failureThreshold: 60
          image: nvcr.io/nvidian/nim-llm-dev/sglang-runtime:hzhou-0811-1
          workingDir: /workspace/components/backends/sglang
          args:
            - "python3"
            - "-m"
            - "dynamo.sglang.decode_worker"
            - "--model-path"
            - "deepseek-ai/DeepSeek-R1-Distill-Llama-8B"
            - "--served-model-name"
            - "deepseek-ai/DeepSeek-R1-Distill-Llama-8B"
            - "--page-size"
            - "16"
            - "--tp"
            - "1"
            - "--trust-remote-code"
            - "--skip-tokenizer-init"
            - "--disaggregation-mode"
            - "decode"
            - "--disaggregation-transfer-backend"
            - "nixl"
    SGLangPrefillWorker:
      dynamoNamespace: dynamo
      envFromSecret: hf-token-secret
      livenessProbe:
        httpGet:
          path: /live
          port: 9090
        periodSeconds: 5
        timeoutSeconds: 30
        failureThreshold: 1
      readinessProbe:
        httpGet:
          path: /health
          port: 9090
        periodSeconds: 10
        timeoutSeconds: 30
        failureThreshold: 60
      componentType: worker
      replicas: 2
      resources:
        requests:
          cpu: "10"
          memory: "20Gi"
          gpu: "1"
        limits:
          cpu: "32"
          memory: "80Gi"
          gpu: "1"
      envs:
        - name: DYN_SYSTEM_ENABLED
          value: "true"
        - name: DYN_SYSTEM_USE_ENDPOINT_HEALTH_STATUS
          value: "[\"generate\"]"
        - name: DYN_SYSTEM_PORT
          value: "9090"
      extraPodSpec:
        mainContainer:
          startupProbe:
            httpGet:
              path: /health
              port: 9090
            periodSeconds: 10
            failureThreshold: 60
          image: nvcr.io/nvidian/nim-llm-dev/sglang-runtime:hzhou-0811-1
          workingDir: /workspace/components/backends/sglang
          args:
            - "python3"
            - "-m"
            - "dynamo.sglang.worker"
            - "--model-path"
            - "deepseek-ai/DeepSeek-R1-Distill-Llama-8B"
            - "--served-model-name"
            - "deepseek-ai/DeepSeek-R1-Distill-Llama-8B"
            - "--page-size"
            - "16"
            - "--tp"
            - "1"
            - "--trust-remote-code"
            - "--skip-tokenizer-init"
            - "--disaggregation-mode"
            - "prefill"
            - "--disaggregation-transfer-backend"
            - "nixl"