Benefits
| Benefit | Description |
|---|---|
| Unified storage | All cloud providers in one place |
| Long-term retention | Keep data as long as you need |
| Powerful querying | PromQL for metrics, LogQL for logs |
| Unified alerting | One alerting system across all clouds |
How data flows
| Method | Description | Best for |
|---|---|---|
| Scrape Jobs | Grafana Cloud pulls via API. | Quick start, evaluation, Azure production |
| Metric Streams | Cloud pushes in real-time. | AWS production (recommended) |
| Grafana Alloy | Universal collector agent | Multi-cloud, complex processing |
Script
Here’s what you get when you store your cloud data in Grafana Cloud.
First, unification. Imagine an incident where something in AWS is affecting a service in Azure. With stored data, you can write a single query that correlates metrics from both clouds.
Second, you control retention. Keep data for months or years, not just the 15-93 days cloud providers offer. Third, consistent query languages. Learn PromQL once and use it everywhere. Fourth, unified alerting across all your clouds.
Now, how does data actually flow? There are a few collection mechanisms. Scrape jobs let Grafana Cloud pull data from your cloud provider’s API on a schedule. Nothing to deploy, great for evaluation or simple setups.
For AWS specifically, we recommend metric streams for production. They push data in near real-time, have a better cost profile at scale, and give you lower latency for alerting.
Azure’s serverless scrape jobs work great for production there. Grafana Alloy is a universal collector you deploy when you need maximum flexibility for multi-cloud or complex processing needs.