# Metric Monitoring

### Popular Option: Prometheus

<figure><img src="https://3885248957-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FoE4wMO1dMVDOGDjh0En7%2Fuploads%2FZtEumeQL5Hxs5tGC678G%2Fimage.png?alt=media&#x26;token=816bbbc9-5df3-47a4-9d1a-337136047739" alt=""><figcaption></figcaption></figure>

### 4 Key components:

* The TSDB (time series database) actually stores the billions of individual measurements of things like page load time
* The "retrieval" section takes data from various places, like parsing them from log messages or measuring how long individual jobs took, and puts them in the database
* The alert manager sends notifications to the relevant people when INCIDENT
* The web UI is primary used to understand the metrics alerted someone has been alerted of a fault