docs: hello world and vllm process docs (#525)

examples/hello_world/README.md

@@ -15,8 +15,17 @@ See the License for the specific language governing permissions and
 limitations under the License.
 -->
 
+# Hello World Example
+
 ## Overview
 
+This example demonstrates the basic concepts of Dynamo by creating a simple multi-service pipeline. It shows how to:
+
+1. Create and connect multiple Dynamo services
+2. Pass data between services using Dynamo's runtime
+3. Set up a simple HTTP API endpoint
+4. Deploy and interact with a Dynamo service graph
+
 Pipeline Architecture:
 
 ```
@@ -38,16 +47,35 @@ Users/Clients (HTTP)
 └─────────────┘
 ```
 
+## Component Descriptions
+
+### Frontend Service
+- Serves as the entry point for external HTTP requests
+- Exposes a `/generate` HTTP API endpoint that clients can call
+- Processes incoming text and passes it to the Middle service
+
+### Middle Service
+- Acts as an intermediary service in the pipeline
+- Receives requests from the Frontend
+- Appends "-mid" to the text and forwards it to the Backend
+
+### Backend Service
+- Functions as the final service in the pipeline
+- Processes requests from the Middle service
+- Appends "-back" to the text and yields tokens
+
+## Running the Example
 
-## Unified serve
-1. Launch all three services using a single command -
+1. Launch all three services using a single command:
 
 ```bash
 cd /workspace/examples/hello_world
 dynamo serve hello_world:Frontend
 ```
 
-2. Send request to frontend using curl -
+The `dynamo serve` command deploys the entire service graph, automatically handling the dependencies between Frontend, Middle, and Backend services.
+
+2. Send request to frontend using curl:
 
 ```bash
 curl -X 'POST' \
@@ -58,3 +86,16 @@ curl -X 'POST' \
   "text": "test"
 }'
 ```
+
+## Expected Output
+
+When you send the request with "test" as input, the response will show how the text flows through each service:
+
+```
+Frontend: Middle: Backend: test-mid-back
+```
+
+This demonstrates how:
+1. The Frontend receives "test"
+2. The Middle service adds "-mid" to create "test-mid"
+3. The Backend service adds "-back" to create "test-mid-back"
\ No newline at end of file

examples/hello_world/hello_world.py

@@ -13,10 +13,14 @@
 # See the License for the specific language governing permissions and
 # limitations under the License.
 
+import logging
+
 from pydantic import BaseModel
 
 from dynamo.sdk import DYNAMO_IMAGE, api, depends, dynamo_endpoint, service
 
+logger = logging.getLogger(__name__)
+
 """
 Pipeline Architecture:
 
@@ -48,32 +52,27 @@ class ResponseType(BaseModel):
 
 
 @service(
-    resources={"cpu": "2"},
-    traffic={"timeout": 30},
     dynamo={
         "enabled": True,
         "namespace": "inference",
     },
-    workers=3,
     image=DYNAMO_IMAGE,
 )
 class Backend:
     def __init__(self) -> None:
-        print("Starting backend")
+        logger.info("Starting backend")
 
     @dynamo_endpoint()
     async def generate(self, req: RequestType):
         """Generate tokens."""
         req_text = req.text
-        print(f"Backend received: {req_text}")
+        logger.info(f"Backend received: {req_text}")
         text = f"{req_text}-back"
         for token in text.split():
             yield f"Backend: {token}"
 
 
 @service(
-    resources={"cpu": "2"},
-    traffic={"timeout": 30},
     dynamo={"enabled": True, "namespace": "inference"},
     image=DYNAMO_IMAGE,
 )
@@ -81,23 +80,21 @@ class Middle:
     backend = depends(Backend)
 
     def __init__(self) -> None:
-        print("Starting middle")
+        logger.info("Starting middle")
 
     @dynamo_endpoint()
     async def generate(self, req: RequestType):
         """Forward requests to backend."""
         req_text = req.text
-        print(f"Middle received: {req_text}")
+        logger.info(f"Middle received: {req_text}")
         text = f"{req_text}-mid"
         next_request = RequestType(text=text).model_dump_json()
         async for response in self.backend.generate(next_request):
-            print(f"Middle received response: {response}")
+            logger.info(f"Middle received response: {response}")
             yield f"Middle: {response}"
 
 
 @service(
-    resources={"cpu": "1"},
-    traffic={"timeout": 60},
     image=DYNAMO_IMAGE,
 )  # Regular HTTP API
 class Frontend:

examples/llm/README.md

@@ -157,9 +157,8 @@ See [multinode-examples.md](multinode-examples.md) for more details.
 
 ### Close deployment
 
-Kill all dynamo processes managed by circusd.
+> [!IMPORTANT]
+> We are aware of an issue where vLLM subprocesses might not be killed when `ctrl-c` is pressed.
+> We are working on addressing this. Relevant vLLM issues can be found [here](https://github.com/vllm-project/vllm/pull/8492) and [here](https://github.com/vllm-project/vllm/issues/6219#issuecomment-2439257824).
 
-```
-ctrl-c
+To stop the serve, you can press `ctrl-c` which will kill the different components. In order to kill the remaining vLLM subprocesses you can run `nvidia-smi` and `kill -9` the remaining processes or run `pkill python3` from inside of the container.
\ No newline at end of file
-pkill python3
-```