> For the complete documentation index, see [llms.txt](https://docs.cloudeka.ai/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.cloudeka.ai/guidance-for-enterprise/deka-gpu/deka-gpu-metric-collection/metric-aggregator.md).

# Metric Aggregator

Metric Aggregator uses Prometheus as the metric aggregator to scrape and store metrics from annotated pods. The Prometheus server is pre-configured to discover and scrape metrics from pods with the specified annotations. Below are the steps to access and query metrics:

### 1. Access Prometheus UI

Open the Prometheus web interface using the provided URL and provided credentials.

* Check target status at `status -> Targets`
* Verify that the pod is listed and the status is "UP".

### 2. Explore Metrics

Use the "Graph" tab to enter PromQL queries and visualize metrics.

* PromQL query example:

```yaml
vllm:gpu_cache_usage_perc{job="prod-cluster-1", model_name="meta-llama/Llama-3.2-3B- Instruct"}
```