Meeting the operational challenge

Wed, 17 Jun 2026 12:54:46 -0500

Cut the cost of running your own stack

Running a self-managed stack is real engineering work, separate from using the data:

Capacity planning, patching, upgrades, and on-call for the observability backend itself
Security reviews and version maintenance
Engineering hours that don’t show up on any vendor invoice

With Grafana Cloud, you can manage these operational challenges:

Cardinality: Every new service and label combination adds unique metric series, and high cardinality (metrics with many unique label combinations) drives memory and cost.
- The Cloud Metrics database is built for high-cardinality workloads, no need to resize and tune the backend to keep up.
- Adaptive Metrics automatically combines any series that is not used by dashboards, alerts, or queries.
Ingestion volume: Logs and traces arrive faster as traffic rises. The built-in databases Grafana Cloud Logs and Grafana Cloud Traces can scale without you performing capacity planning.
Sharding and scaling: Backend databases run as distributed systems that need sharding (splitting data across nodes), capacity planning, and careful upgrades.

SailPoint

“When we crunched the numbers, Grafana Labs was offering to run everything for cheaper, and it would reduce the load on our engineering team. That was our ‘a-ha’ moment. We spent a lot of time managing Grafana internally. Now we have more time to innovate.”
Paradigm

“We could have hosted Loki internally, which is the beautiful thing about Grafana’s open source tooling. But we chose Grafana Cloud because we’re a small company and a small team, so we didn’t want to manage it ourselves. The engineering costs of hosting it ourselves would be much higher.”
Mux

“Grafana Cloud probably saves us hundreds of engineering hours a year. Our platform engineers don’t have to manage the stack any more, and our product engineers don’t have to work through multiple observability tools.”
Hona

“We considered self-hosting the open source version, but when we calculated the engineering time for setup and maintenance versus Grafana Cloud, Grafana Cloud made much more sense.”

Wed, 17 Jun 2026 12:54:46 -0500

Often, a high-traffic system produces far more traces than you can affordably store and query on your own.

Grafana Cloud Traces runs that backend and storage for you, so you can afford to keep enough trace history to investigate.

No backend to operate: Storage, scaling, sharding, and capacity planning are handled for you.
Adaptive Traces: Healthy requests produce far more traces than you need, but tail sampling keeps what matters and drops the rest. Open source can tail-sample too, but you run the pipeline yourself and redeploy it to change policies. Adaptive Traces is the managed version, where policy changes take effect within about 10 seconds, no redeploy.

Profiling works the same in open source and Grafana Cloud, but the difference is operating it.

Always-on profiling generates continuous data from every instance in your fleet.
Storing and querying that volume affordably is hard to run yourself.

Grafana Cloud Profiles stores and queries that volume for you, so you keep profiling on everywhere.

Uber

Moved from manual, reactive profiling to always-on continuous profiling with Grafana Cloud Profiles.
IG Group

Chose Grafana Cloud specifically for Tempo as their traces store at scale, alongside Mimir for metrics. Built an “observable by default” platform using OpenTelemetry across cloud and on-prem.