fix: update kv-router usage (#1238)

docs/guides/planner_benchmark/benchmark_planner.md

@@ -48,7 +48,7 @@ To measure the performance of dynamo with planner, we start from a 1p1d deployme
 
 ```bash
 cd examples/llm
-dynamo serve graphs.disagg:Frontend -f disagg_1p1d.yml
+dynamo serve graphs.disagg_router:Frontend -f disagg_1p1d.yml
 
 # in terminal 2
 genai-perf profile \
@@ -82,7 +82,7 @@ In this example, we use a fixed 2p2d engine as baseline. Planner provides a `--n
 
 ```bash
 # in terminal 1
-dynamo serve graphs.disagg:Frontend -f disagg_2p2d.yml
+dynamo serve graphs.disagg_router:Frontend -f disagg_2p2d.yml
 
 # in terminal 2
 genai-perf profile --tokenizer deepseek-ai/DeepSeek-R1-Distill-Llama-8B -m deepseek-ai/DeepSeek-R1-Distill-Llama-8B --service-kind openai --endpoint-type chat --url http://localhost:8000 --streaming --input-file payload:sin_b512_t600_rr5.0-20.0-150.0_io3000150-3000150-0.2-0.8-10.jsonl

docs/guides/planner_benchmark/disagg_1p1d.yml

@@ -17,6 +17,7 @@ Common:
   block-size: 64
   max-model-len: 16384
   kv-transfer-config: '{"kv_connector":"DynamoNixlConnector"}'
+  router: kv-load
 
 Frontend:
   served_model_name: deepseek-ai/DeepSeek-R1-Distill-Llama-8B
@@ -24,8 +25,11 @@ Frontend:
   port: 8000
 
 Processor:
-  router: kv-load
-  common-configs: [model, block-size]
+  common-configs: [model, block-size, router]
+
+Router:
+  min-workers: 1
+  common-configs: [model, block-size, router]
 
 VllmWorker:
   remote-prefill: true

docs/guides/planner_benchmark/disagg_2p2d.yaml

@@ -17,6 +17,7 @@ Common:
   block-size: 64
   max-model-len: 16384
   kv-transfer-config: '{"kv_connector":"DynamoNixlConnector"}'
+  router: kv-load
 
 Frontend:
   served_model_name: deepseek-ai/DeepSeek-R1-Distill-Llama-8B
@@ -25,7 +26,11 @@ Frontend:
 
 Processor:
   router: kv-load
-  common-configs: [model, block-size]
+  common-configs: [model, block-size, router]
+
+Router:
+  min-workers: 1
+  common-configs: [model, block-size, router]
 
 VllmWorker:
   remote-prefill: true