Question 1

What is Gigapipe?

Accepted Answer

Gigapipe is an open source, OpenTelemetry-native observability backend that stores logs, metrics, traces and profiles in ClickHouse. It ingests OTLP over gRPC and HTTP, along with the Loki, Prometheus, Tempo, Zipkin, Pyroscope, InfluxDB, Elastic and Datadog protocols, and answers LogQL, PromQL, TraceQL and FlameQL queries over the same data. It is polyglot by design: one store, many protocols, so the agents and dashboards you already run keep working.

Question 2

Does Gigapipe support the Loki, Prometheus and Tempo APIs?

Accepted Answer

Yes. Gigapipe implements the native Loki, Prometheus, Tempo and Pyroscope APIs for both writing and reading, so it is a drop-in target for anything already speaking them. In practice that means Promtail, Grafana Alloy, Vector, Fluent Bit, Telegraf, Filebeat and the OpenTelemetry Collector can all write to it unchanged, and Grafana or Perses can read from it through their built-in datasources with no plugin to install.

Question 3

Is Gigapipe an OpenTelemetry-native backend?

Accepted Answer

Yes. OTLP is a first-class ingest path over both gRPC and HTTP, for logs, metrics, traces and profiles, and OpenTelemetry semantic conventions and high-cardinality attributes are preserved rather than flattened into a proprietary schema. An unmodified OpenTelemetry Collector exports to Gigapipe with a standard exporter block. Gigapipe is listed on the OpenTelemetry vendor registry.

Question 4

Can I self-host Gigapipe?

Accepted Answer

Yes. Gigapipe is AGPL-3.0 and runs from a container, a Kubernetes deployment or a single binary, backed by your own ClickHouse and object storage. The open source release is the same engine as the hosted offering — there is no open-core split where the useful parts are held back.

Question 5

Why is Gigapipe built on ClickHouse?

Accepted Answer

ClickHouse is a columnar database built for exactly the shape of an observability workload: high-volume append-only writes, aggressive compression, and analytical queries over wide time ranges. Storing every signal in one ClickHouse cluster is also what makes cross-signal correlation cheap, and it means your telemetry sits in a database you can query directly with SQL rather than a black box you can only reach through a vendor UI.

Question 6

How does Gigapipe compare to Datadog or Grafana Cloud?

Accepted Answer

The functional difference is protocol compatibility and where your data lives: Gigapipe speaks the open ingest and query APIs, runs on infrastructure you control, and stores data in your own ClickHouse and object storage. The commercial difference is the pricing shape. Per-GB and per-host billing means your bill scales with how much you observe, which pushes teams to sample and drop data precisely when they need it most. Gigapipe charges by infrastructure size, so ingestion and queries are unlimited within your plan.

Your Observability Stack.
Your data. Your rules.

Polyglot Compatibility

Unified Storage Layer

Flexible Deployment

Open Source & Vendor-Agnostic

Open-source

Pro

Enterprise

About us

Keep learning...

What If You Didn't Have to Migrate to Gigapipe to Find Out If It's Better?

Wait, you can use ALL of Grafana Cloud's AI features for free?!

Your Monitoring Thinks 3 AM Looks Like 3 PM. That's Why You Miss Incidents.

🤖 Your AI Agent Is Burning Money. You Just Can't See It Yet.

Your Observability Stack.Your data. Your rules.