feat: normalize dynamo namespace computation (#5231)

Signed-off-by: Hannah Zhang <hannahz@nvidia.com>

recipes/qwen3-32b/vllm/disagg-kv-router/deploy.yaml

@@ -13,7 +13,6 @@ spec:
   services:
     Frontend:
       componentType: frontend
-      dynamoNamespace: disagg-router-6p-2d
       envs:
         - name: HF_HOME
           value: /home/dynamo/.cache/huggingface
@@ -38,7 +37,6 @@ spec:
       subComponentType: null
     VllmDecodeWorker:
       componentType: worker
-      dynamoNamespace: disagg-router-6p-2d
       envFromSecret: hf-token-secret
       extraPodSpec:
         mainContainer:
@@ -86,7 +84,6 @@ spec:
         useAsCompilationCache: true
     VllmPrefillWorker:
       componentType: worker
-      dynamoNamespace: disagg-router-6p-2d
       envFromSecret: hf-token-secret
       extraPodMetadata:
         annotations:

tests/fault_tolerance/deploy/templates/vllm/moe_agg.yaml

@@ -8,7 +8,6 @@ metadata:
 spec:
   services:
     Frontend:
-      dynamoNamespace: vllm-moe-agg
       componentType: frontend
       replicas: 1
       extraPodSpec:
@@ -16,7 +15,6 @@ spec:
           image: nvcr.io/nvidia/ai-dynamo/vllm-runtime:my-tag
     VllmDecodeWorker:
       envFromSecret: hf-token-secret
-      dynamoNamespace: vllm-moe-agg
       componentType: worker
       replicas: 1
       resources:

tests/fault_tolerance/deploy/templates/vllm/moe_disagg.yaml

@@ -8,7 +8,6 @@ metadata:
 spec:
   services:
     Frontend:
-      dynamoNamespace: vllm-moe-disagg
       componentType: frontend
       replicas: 1
       extraPodSpec:
@@ -17,7 +16,6 @@ spec:
         mainContainer:
           image: nvcr.io/nvidia/ai-dynamo/vllm-runtime:my-tag
     VllmDecodeWorker:
-      dynamoNamespace: vllm-moe-disagg
       envFromSecret: hf-token-secret
       componentType: worker
       subComponentType: decode
@@ -83,7 +81,6 @@ spec:
             - --no-enable-prefix-caching
             - --enforce-eager
     VllmPrefillWorker:
-      dynamoNamespace: vllm-moe-disagg
       envFromSecret: hf-token-secret
       componentType: worker
       subComponentType: prefill

tests/planner/perf_test_configs/agg_8b.yaml

@@ -8,7 +8,6 @@ metadata:
 spec:
   services:
     Frontend:
-      dynamoNamespace: vllm-agg
       componentType: main
       replicas: 1
       livenessProbe:
@@ -46,7 +45,6 @@ spec:
           args:
             - "python3 -m dynamo.frontend --http-port 8000"
     VllmDecodeWorker:
-      dynamoNamespace: vllm-agg
       envFromSecret: hf-token-secret
       componentType: worker
       replicas: 1

tests/planner/perf_test_configs/disagg_8b_2p2d.yaml

@@ -8,7 +8,6 @@ metadata:
 spec:
   services:
     Frontend:
-      dynamoNamespace: vllm-disagg
       componentType: main
       replicas: 1
       livenessProbe:
@@ -46,7 +45,6 @@ spec:
           args:
             - "python3 -m dynamo.frontend --http-port 8000 --kv-cache-block-size 128"
     VllmDecodeWorker:
-      dynamoNamespace: vllm-disagg
       envFromSecret: hf-token-secret
       componentType: worker
       replicas: 2
@@ -94,7 +92,6 @@ spec:
           args:
             - "python3 -m dynamo.vllm --model nvidia/Llama-3.1-8B-Instruct-FP8 --no-enable-prefix-caching --block-size 128 2>&1 | tee /tmp/vllm.log"
     VllmPrefillWorker:
-      dynamoNamespace: vllm-disagg
       envFromSecret: hf-token-secret
       componentType: worker
       replicas: 2

tests/planner/perf_test_configs/disagg_8b_3p1d.yaml

@@ -8,7 +8,6 @@ metadata:
 spec:
   services:
     Frontend:
-      dynamoNamespace: vllm-disagg
       componentType: main
       replicas: 1
       livenessProbe:
@@ -46,7 +45,6 @@ spec:
           args:
             - "python3 -m dynamo.frontend --http-port 8000 --kv-cache-block-size 128"
     VllmDecodeWorker:
-      dynamoNamespace: vllm-disagg
       envFromSecret: hf-token-secret
       componentType: worker
       replicas: 1
@@ -94,7 +92,6 @@ spec:
           args:
             - "python3 -m dynamo.vllm --model nvidia/Llama-3.1-8B-Instruct-FP8 --no-enable-prefix-caching --block-size 128 2>&1 | tee /tmp/vllm.log"
     VllmPrefillWorker:
-      dynamoNamespace: vllm-disagg
       envFromSecret: hf-token-secret
       componentType: worker
       replicas: 3

tests/planner/perf_test_configs/disagg_8b_planner.yaml

@@ -11,7 +11,6 @@ spec:
       value: '{"Prometheus":{"global":{"scrape_interval":"5s"},"scrape_configs":[{"job_name":"prometheus","static_configs":[{"targets":["localhost:8000"]}]},{"job_name":"frontend","static_configs":[{"targets":["vllm-disagg-planner-frontend:8000"]}]}]}}'
   services:
     Frontend:
-      dynamoNamespace: vllm-disagg-planner
       componentType: main
       replicas: 1
       livenessProbe:
@@ -49,7 +48,6 @@ spec:
           args:
             - "python3 -m dynamo.frontend --http-port 8000 --kv-cache-block-size 128 --router-mode kv --kv-overlap-score-weight 0.0 --router-temperature 0.0 --no-kv-events"
     Planner:
-      dynamoNamespace: vllm-disagg-planner
       envFromSecret: hf-token-secret
       componentType: planner
       replicas: 1
@@ -94,7 +92,6 @@ spec:
               --prometheus-port=9085
               --no-correction
     VllmDecodeWorker:
-      dynamoNamespace: vllm-disagg-planner
       envFromSecret: hf-token-secret
       componentType: worker
       subComponentType: decode
@@ -149,7 +146,6 @@ spec:
             - --block-size
             - "128"
     VllmPrefillWorker:
-      dynamoNamespace: vllm-disagg-planner
       envFromSecret: hf-token-secret
       componentType: worker
       subComponentType: prefill

tests/planner/perf_test_configs/disagg_8b_tp2.yaml

@@ -8,7 +8,6 @@ metadata:
 spec:
   services:
     Frontend:
-      dynamoNamespace: vllm-disagg
       componentType: main
       replicas: 1
       livenessProbe:
@@ -46,7 +45,6 @@ spec:
           args:
             - "python3 -m dynamo.frontend --http-port 8000 --kv-cache-block-size 128"
     VllmDecodeWorker:
-      dynamoNamespace: vllm-disagg
       envFromSecret: hf-token-secret
       componentType: worker
       replicas: 1
@@ -94,7 +92,6 @@ spec:
           args:
             - "python3 -m dynamo.vllm --model nvidia/Llama-3.1-8B-Instruct-FP8 --no-enable-prefix-caching --block-size 128 --tensor-parallel-size 2 2>&1 | tee /tmp/vllm.log"
     VllmPrefillWorker:
-      dynamoNamespace: vllm-disagg
       envFromSecret: hf-token-secret
       componentType: worker
       replicas: 1

tests/planner/profiling_results/H200_TP1P_TP1D/disagg.yaml

@@ -8,7 +8,6 @@ metadata:
 spec:
   services:
     Frontend:
-      dynamoNamespace: vllm-disagg
       componentType: main
       replicas: 1
       livenessProbe:
@@ -46,7 +45,6 @@ spec:
           args:
             - "python3 -m dynamo.frontend --http-port 8000"
     VllmDecodeWorker:
-      dynamoNamespace: vllm-disagg
       envFromSecret: hf-token-secret
       componentType: worker
       replicas: 1
@@ -94,7 +92,6 @@ spec:
           args:
             - "python3 -m dynamo.vllm --model nvidia/Llama-3.1-8B-Instruct-FP8  2>&1 | tee /tmp/vllm.log"
     VllmPrefillWorker:
-      dynamoNamespace: vllm-disagg
       envFromSecret: hf-token-secret
       componentType: worker
       replicas: 1