feat: Add model label for vllm backend metrics (#2474)

Co-authored-by: Keiven Chang <[email protected]>
Signed-off-by: Hannah Zhang <[email protected]>

Author: 2 people

components/backends/vllm/src/dynamo/vllm/main.py

@@ -150,8 +150,14 @@ async def init_prefill(runtime: DistributedRuntime, config: Config):
             #     (temp reason): we don't support re-routing prefill requests
             #     (long-term reason): prefill engine should pull from a global queue so there is
             #                         only a few in-flight requests that can be quickly finished
-            generate_endpoint.serve_endpoint(handler.generate, graceful_shutdown=True),
-            clear_endpoint.serve_endpoint(handler.clear_kv_blocks),
+            generate_endpoint.serve_endpoint(
+                handler.generate,
+                graceful_shutdown=True,
+                metrics_labels=[("model", config.model)],
+            ),
+            clear_endpoint.serve_endpoint(
+                handler.clear_kv_blocks, metrics_labels=[("model", config.model)]
+            ),
         )
     except Exception as e:
         logger.error(f"Failed to serve endpoints: {e}")
@@ -178,7 +184,11 @@ async def init(runtime: DistributedRuntime, config: Config):
         .client()
     )
 
-    factory = StatLoggerFactory(component, config.engine_args.data_parallel_rank or 0)
+    factory = StatLoggerFactory(
+        component,
+        config.engine_args.data_parallel_rank or 0,
+        metrics_labels=[("model", config.model)],
+    )
     engine_client, vllm_config, default_sampling_params = setup_vllm_engine(
         config, factory
     )
@@ -239,8 +249,14 @@ async def init(runtime: DistributedRuntime, config: Config):
         await asyncio.gather(
             # for decode, we want to transfer the in-flight requests to other decode engines,
             # because waiting them to finish can take a long time for long OSLs
-            generate_endpoint.serve_endpoint(handler.generate, graceful_shutdown=False),
-            clear_endpoint.serve_endpoint(handler.clear_kv_blocks),
+            generate_endpoint.serve_endpoint(
+                handler.generate,
+                graceful_shutdown=False,
+                metrics_labels=[("model", config.model)],
+            ),
+            clear_endpoint.serve_endpoint(
+                handler.clear_kv_blocks, metrics_labels=[("model", config.model)]
+            ),
         )
     except Exception as e:
         logger.error(f"Failed to serve endpoints: {e}")

components/backends/vllm/src/dynamo/vllm/publisher.py

@@ -1,7 +1,7 @@
 # SPDX-FileCopyrightText: Copyright (c) 2025 NVIDIA CORPORATION & AFFILIATES. All rights reserved.
 # SPDX-License-Identifier: Apache-2.0
 
-from typing import Optional
+from typing import List, Optional, Tuple
 
 from vllm.config import VllmConfig
 from vllm.v1.metrics.loggers import StatLoggerBase
@@ -36,9 +36,16 @@ def log_engine_initialized(self):
 class DynamoStatLoggerPublisher(StatLoggerBase):
     """Stat logger publisher. Wrapper for the WorkerMetricsPublisher to match the StatLoggerBase interface."""
 
-    def __init__(self, component: Component, dp_rank: int) -> None:
+    def __init__(
+        self,
+        component: Component,
+        dp_rank: int,
+        metrics_labels: Optional[List[Tuple[str, str]]] = None,
+    ) -> None:
         self.inner = WorkerMetricsPublisher()
-        self.inner.create_endpoint(component)
+        # Use labels directly for the new create_endpoint signature
+        metrics_labels = metrics_labels or []
+        self.inner.create_endpoint(component, metrics_labels)
         self.dp_rank = dp_rank
         self.num_gpu_block = 1
         self.request_total_slots = 1
@@ -129,15 +136,23 @@ def log_engine_initialized(self) -> None:
 class StatLoggerFactory:
     """Factory for creating stat logger publishers. Required by vLLM."""
 
-    def __init__(self, component: Component, dp_rank: int = 0) -> None:
+    def __init__(
+        self,
+        component: Component,
+        dp_rank: int = 0,
+        metrics_labels: Optional[List[Tuple[str, str]]] = None,
+    ) -> None:
         self.component = component
         self.created_logger: Optional[DynamoStatLoggerPublisher] = None
         self.dp_rank = dp_rank
+        self.metrics_labels = metrics_labels or []
 
     def create_stat_logger(self, dp_rank: int) -> StatLoggerBase:
         if self.dp_rank != dp_rank:
             return NullStatLogger()
-        logger = DynamoStatLoggerPublisher(self.component, dp_rank)
+        logger = DynamoStatLoggerPublisher(
+            self.component, dp_rank, metrics_labels=self.metrics_labels
+        )
         self.created_logger = logger
 
         return logger

examples/multimodal/components/worker.py

@@ -140,7 +140,9 @@ def setup_vllm_engine(self, component: Component, endpoint: Endpoint):
 
         # Create vLLM engine with metrics logger and KV event publisher attached
         self.stats_logger = StatLoggerFactory(
-            component, self.engine_args.data_parallel_rank or 0
+            component,
+            self.engine_args.data_parallel_rank or 0,
+            metrics_labels=[("model", self.engine_args.model)],
         )
         self.engine_client = AsyncLLM.from_vllm_config(
             vllm_config=vllm_config,
@@ -353,7 +355,9 @@ async def generate(self, request: vLLMMultimodalRequest):
                 extra_args.pop("serialized_request", None)
                 decode_request.sampling_params.extra_args = extra_args
                 logger.debug("Decode request: %s", decode_request)
-                async for decode_response in await self.decode_worker_client.round_robin(
+                async for (
+                    decode_response
+                ) in await self.decode_worker_client.round_robin(
                     decode_request.model_dump_json()
                 ):
                     output = MyRequestOutput.model_validate_json(decode_response.data())

lib/bindings/python/rust/lib.rs

@@ -513,19 +513,24 @@ impl Component {
 
 #[pymethods]
 impl Endpoint {
-    #[pyo3(signature = (generator, graceful_shutdown = true))]
+    #[pyo3(signature = (generator, graceful_shutdown = true, metrics_labels = None))]
     fn serve_endpoint<'p>(
         &self,
         py: Python<'p>,
         generator: PyObject,
         graceful_shutdown: Option<bool>,
+        metrics_labels: Option<Vec<(String, String)>>,
     ) -> PyResult<Bound<'p, PyAny>> {
         let engine = Arc::new(engine::PythonAsyncEngine::new(
             generator,
             self.event_loop.clone(),
         )?);
         let ingress = JsonServerStreamingIngress::for_engine(engine).map_err(to_pyerr)?;
-        let builder = self.inner.endpoint_builder().handler(ingress);
+        let builder = self
+            .inner
+            .endpoint_builder()
+            .metrics_labels(metrics_labels)
+            .handler(ingress);
         let graceful_shutdown = graceful_shutdown.unwrap_or(true);
         pyo3_async_runtimes::tokio::future_into_py(py, async move {
             builder

lib/bindings/python/rust/llm/kv.rs

@@ -55,17 +55,35 @@ impl WorkerMetricsPublisher {
         })
     }
 
-    #[pyo3(signature = (component))]
+    #[pyo3(signature = (component, metrics_labels = None))]
     fn create_endpoint<'p>(
         &self,
         py: Python<'p>,
         component: Component,
+        metrics_labels: Option<Vec<(String, String)>>,
     ) -> PyResult<Bound<'p, PyAny>> {
         let rs_publisher = self.inner.clone();
         let rs_component = component.inner.clone();
         pyo3_async_runtimes::tokio::future_into_py(py, async move {
+            // Convert Python labels to Option<&[(&str, &str)]> expected by Rust API
+            let metrics_labels_ref: Option<Vec<(&str, &str)>> =
+                if let Some(metrics_labels) = metrics_labels.as_ref() {
+                    if metrics_labels.is_empty() {
+                        None
+                    } else {
+                        Some(
+                            metrics_labels
+                                .iter()
+                                .map(|(k, v)| (k.as_str(), v.as_str()))
+                                .collect(),
+                        )
+                    }
+                } else {
+                    None
+                };
+
             rs_publisher
-                .create_endpoint(rs_component)
+                .create_endpoint(rs_component, metrics_labels_ref.as_deref())
                 .await
                 .map_err(to_pyerr)?;
             Ok(())

lib/llm/src/kv_router/publisher.rs

@@ -498,7 +498,11 @@ impl WorkerMetricsPublisher {
         self.tx.send(metrics)
     }
 
-    pub async fn create_endpoint(&self, component: Component) -> Result<()> {
+    pub async fn create_endpoint(
+        &self,
+        component: Component,
+        metrics_labels: Option<&[(&str, &str)]>,
+    ) -> Result<()> {
         let mut metrics_rx = self.rx.clone();
         let handler = Arc::new(KvLoadEndpointHandler::new(metrics_rx.clone()));
         let handler = Ingress::for_engine(handler)?;
@@ -514,13 +518,20 @@ impl WorkerMetricsPublisher {
 
         self.start_nats_metrics_publishing(component.namespace().clone(), worker_id);
 
+        let metrics_labels = metrics_labels.map(|v| {
+            v.iter()
+                .map(|(k, v)| (k.to_string(), v.to_string()))
+                .collect::<Vec<_>>()
+        });
+
         component
             .endpoint(KV_METRICS_ENDPOINT)
             .endpoint_builder()
             .stats_handler(move |_| {
                 let metrics = metrics_rx.borrow_and_update().clone();
                 serde_json::to_value(&*metrics).unwrap()
             })
+            .metrics_labels(metrics_labels)
             .handler(handler)
             .start()
             .await

lib/llm/src/mocker/engine.rs

@@ -189,7 +189,7 @@ impl MockVllmEngine {
             tokio::spawn({
                 let publisher = metrics_publisher.clone();
                 async move {
-                    if let Err(e) = publisher.create_endpoint(comp.clone()).await {
+                    if let Err(e) = publisher.create_endpoint(comp.clone(), None).await {
                         tracing::error!("Metrics endpoint failed: {e}");
                     }
                 }

lib/runtime/examples/system_metrics/src/lib.rs

@@ -16,7 +16,6 @@ use std::sync::Arc;
 pub const DEFAULT_NAMESPACE: &str = "dyn_example_namespace";
 pub const DEFAULT_COMPONENT: &str = "dyn_example_component";
 pub const DEFAULT_ENDPOINT: &str = "dyn_example_endpoint";
-pub const DEFAULT_MODEL_NAME: &str = "dyn_example_model";
 
 /// Stats structure returned by the endpoint's stats handler
 #[derive(serde::Serialize, serde::Deserialize, Debug, Clone)]

lib/runtime/src/component/endpoint.rs

@@ -41,6 +41,10 @@ pub struct EndpointConfig {
     #[builder(default, private)]
     _stats_handler: Option<EndpointStatsHandler>,
 
+    /// Additional labels for metrics
+    #[builder(default, setter(into))]
+    metrics_labels: Option<Vec<(String, String)>>,
+
     /// Whether to wait for inflight requests to complete during shutdown
     #[builder(default = "true")]
     graceful_shutdown: bool,
@@ -59,7 +63,7 @@ impl EndpointConfigBuilder {
     }
 
     pub async fn start(self) -> Result<()> {
-        let (endpoint, lease, handler, stats_handler, graceful_shutdown) =
+        let (endpoint, lease, handler, stats_handler, metrics_labels, graceful_shutdown) =
             self.build_internal()?.dissolve();
         let lease = lease.or(endpoint.drt().primary_lease());
         let lease_id = lease.as_ref().map(|l| l.id()).unwrap_or(0);
@@ -74,8 +78,11 @@ impl EndpointConfigBuilder {
         // acquire the registry lock
         let registry = endpoint.drt().component_registry.inner.lock().await;
 
+        let metrics_labels: Option<Vec<(&str, &str)>> = metrics_labels
+            .as_ref()
+            .map(|v| v.iter().map(|(k, v)| (k.as_str(), v.as_str())).collect());
         // Add metrics to the handler. The endpoint provides additional information to the handler.
-        handler.add_metrics(&endpoint)?;
+        handler.add_metrics(&endpoint, metrics_labels.as_deref())?;
 
         // get the group
         let group = registry

lib/runtime/src/pipeline/network.rs

@@ -300,8 +300,12 @@ impl<Req: PipelineIO + Sync, Resp: PipelineIO> Ingress<Req, Resp> {
             .map_err(|_| anyhow::anyhow!("Segment already set"))
     }
 
-    pub fn add_metrics(&self, endpoint: &crate::component::Endpoint) -> Result<()> {
-        let metrics = WorkHandlerMetrics::from_endpoint(endpoint)
+    pub fn add_metrics(
+        &self,
+        endpoint: &crate::component::Endpoint,
+        metrics_labels: Option<&[(&str, &str)]>,
+    ) -> Result<()> {
+        let metrics = WorkHandlerMetrics::from_endpoint(endpoint, metrics_labels)
             .map_err(|e| anyhow::anyhow!("Failed to create work handler metrics: {}", e))?;
 
         self.metrics
@@ -345,7 +349,11 @@ pub trait PushWorkHandler: Send + Sync {
     async fn handle_payload(&self, payload: Bytes) -> Result<(), PipelineError>;
 
     /// Add metrics to the handler
-    fn add_metrics(&self, endpoint: &crate::component::Endpoint) -> Result<()>;
+    fn add_metrics(
+        &self,
+        endpoint: &crate::component::Endpoint,
+        metrics_labels: Option<&[(&str, &str)]>,
+    ) -> Result<()>;
 }
 
 /*