Amazon Managed Grafana
Amazon Managed Grafana runs Grafana as a service. It is a visualisation and alerting layer, not a data store: it queries metrics, logs and traces where they already live and draws them on one dashboard. That is its whole reason for existing — a single view over sources that would otherwise each have their own console.
Workspaces
Section titled “Workspaces”A workspace is a logically isolated Grafana server. You create one, and AWS handles provisioning, setup, scaling, patching and maintenance. There is no instance to size and no Grafana upgrade to run.
Workspaces are the unit of isolation, so they are also the natural boundary for separating environments or teams.
Data sources
Section titled “Data sources”Managed Grafana integrates with the AWS services that hold operational data — Amazon CloudWatch, Amazon Managed Service for Prometheus, Amazon OpenSearch Service, AWS X-Ray, AWS IoT SiteWise and Amazon Timestream — and includes a permission-provisioning feature that sets up the IAM access for a supported AWS source rather than leaving you to write the policy.
Beyond AWS it supports the wide range of open-source, third-party and other cloud data sources that Grafana is known for. That breadth is the argument for choosing it over CloudWatch dashboards: if half your telemetry is in AWS and half is somewhere else, this is where the two meet.
One name on that list needs a caveat: Amazon Timestream for LiveAnalytics has been closed to new customers since 20 June 2025. It remains a supported data source for the accounts already using it; new time-series work goes to Amazon Timestream for InfluxDB or to Amazon Managed Service for Prometheus.
Authentication and governance
Section titled “Authentication and governance”User authentication goes through an identity provider — any that supports SAML 2.0, or AWS IAM Identity Center. Single sign-on, data access control and audit reporting are built in, which is what makes it viable where a self-hosted Grafana would need a compliance case of its own.
Pricing
Section titled “Pricing”Managed Grafana is priced per active user licence per workspace. An active user is one who has signed in to the workspace, or made an API request against it, at least once during the monthly billing cycle — so the cost tracks actual use rather than the size of the directory. Licences are priced separately for editors (including administrators) and viewers, and Grafana Enterprise plugins carry their own per-user charge on top.
Two consequences worth knowing before you create workspaces freely. Every workspace needs at least one editor licence to be managed and signed into, charged whether or not anyone signs in — so a per-team or per-environment workspace sprawl has a floor cost per workspace. And because the licence is per workspace, a person who works across three workspaces is billed three times. Check the pricing page for current rates.
Regional availability
Section titled “Regional availability”The service is in twelve Regions, not every Region: US East (N. Virginia, Ohio), US West (Oregon), Europe (Ireland, London, Frankfurt), Asia Pacific (Tokyo, Seoul, Singapore, Sydney) and both AWS GovCloud Regions. Confirm against the current list before planning a deployment. This constrains where the workspace lives, not what it can see — a workspace queries data sources in other Regions and other accounts perfectly well, so the absence of a Region from this list is rarely a blocker.
Typical uses
Section titled “Typical uses”- Container and Kubernetes metrics, usually over Amazon Managed Service for Prometheus.
- IoT and industrial telemetry from IoT SiteWise.
- Incident troubleshooting, where correlating a metric spike with logs and traces in one place is faster than moving between three consoles.