---
title: "When to connect to a self-hosted data source | Grafana Labs"
description: "Learn about observability and how it can help you solve real-world problems"
---

# When to connect to a self-hosted data source

Grafana Cloud data source connections allow you to visualize metric data directly from where it is stored, meaning the data itself is not sent to nor stored in Grafana Cloud. Watch video ->

Consider connecting to a self-managed metrics data source when:

- The cost of sending metric data to Grafana Cloud is too high
- Compliance requirements prevent storing data off-premise
- You need custom configurations, like support for alpha Prometheus features
- Your retention needs exceed the 13-month limit
- Your team wants to handle the management of the metrics environment internally
- The metrics are already stored and managed in another platform, like Amazon Managed Prometheus or Google Cloud Platform Managed Prometheus
- You prefer to continue with your existing non-Prometheus metric data sources, as you don’t want to transition to Prometheus or Mimir and learn PromQL

* * *

### More to explore (optional)

- [What is Prometheus?](/docs/grafana-cloud/introduction/what-is-observability/prometheus/)
- [Send data from a Prometheus instance](/docs/grafana-cloud/send-data/metrics/metrics-prometheus/#send-data-from-a-prometheus-instance)
- [Send data from multiple Prometheus instances](/docs/grafana-cloud/send-data/metrics/metrics-prometheus/#send-data-from-multiple-prometheus-instances)
- [Send data from multiple high-availability Prometheus instances](/docs/grafana-cloud/send-data/metrics/metrics-prometheus/#send-data-from-multiple-high-availability-prometheus-instances)