Gigapipe: All-In-One Polyglot Observability

August 28, 2026 · View on GitHub

Gigapipe: All-In-One Polyglot Observability

formerly known as qryn

GitHub Repo stars

:rocket: Gigapipe is a lighweight, multi-standard, polyglot observability stack for Logs, Metrics, Traces and Profiling

Polyglot: All-in-one, drop-in compatible with Loki, Prometheus, Tempo, Pyroscope
Familiar: Use stable & popular LogQL, PromQL, TempoQL languages to query and visualize data
Voracious: Ingest using OpenTelemetry, Loki, Prometheus, Tempo, Influx, Datadog, Elastic + more
Versatile: Explore data with qryn's built-in Explorer and CLI or native Grafana datasource compatibility
Secure: Retain total control of data, using ClickHouse, DuckDB or GigAPI with S3 object storage
Independent: Open source, community powered, anti lock-in alternative to vendor controlled stacks

🚀 Get Started

Setup & Deploy gigapipe opensource and get help in our Matrix room :octocat:
Looking for a quick test before installing? Signup for a free trial at gigapipe.com

📚 Documentation

Configuration & Docker — environment variables, Docker quickstart, cross-cluster setup
Profiling API — profiling endpoints, DOT format export, Graphviz visualization
OTLP Profiles Ingestion — OpenTelemetry profiles signal ingestion via /v1development/profiles
OTLP Metrics Ingestion — OpenTelemetry metrics signal ingestion via /v1/metrics (protobuf + JSON)
OTLP over gRPC — OTLP/gRPC receiver for traces, logs, metrics, and profiles
Contributing — development setup, testing, CLA requirement
Full Documentation — hosted docs at gigapipe.com

Features

💡 gigapipe independently implements popular observability standards, protocols and query languages

➡️ Ingestion

📚 OpenTelemetry

⚡ gigapipe is officially integrated with OpenTelemetry supports any log, trace or metric format
Ingested data can be queried using any of the available gigapipe APIs (LogQL, PromQL, TraceQL)

💡 No modifications required to your OpenTelemetry instrumentation!

gigapipe supports native ingestion for Loki, Prometheus, Tempo/Zipkin, Pyroscope and many other protocols
With gigapipe integrators can push and read data using any desired combination of APIs and formats

💡 No opentelemetry or any other middlewayre/proxy required!

⬅️ Query

📚 Loki + LogQL

Any Loki compatible client or application can be used with gigapipe out of the box

⚡ gigapipe implements the Loki API for transparent compatibility with LogQL clients

The Grafana Loki datasource can be used to natively browse and query logs and display extracted timeseries

:tada: No plugins needed
:eye: No Grafana? No problem! Use View

Index Stats & Enhanced Tail

Index statistics: Query stream metadata using /loki/api/v1/index/stats to retrieve stream count, entry count, byte volume, and chunk count. Supports LogQL stream selectors for filtering:

GET /loki/api/v1/index/stats?query={job="app"}&start=<ns>&end=<ns>

Enhanced live-tail: The Tail service now supports limit and start query parameters for controlling tail behavior. The limit parameter caps the number of entries returned, and start sets the starting timestamp (in nanoseconds). WebSocket responses include a dropped_entries field indicating when entries were dropped due to backpressure:

{
  "streams": [...],
  "dropped_entries": []
}

📈 Prometheus + PromQL

Any Prometheus compatible client or application can be used with gigapipe out of the box

⚡ gigapipe implements the Prometheus API for transparent PromQL compatibility using WASM 🏆

The Grafana Prometheus datasource can be used to natively to query metrics and display timeseries

:tada: No plugins needed
:eye: No Grafana? No problem! Use View

🕛 Tempo + TraceQL

⚡ gigapipe implements the Tempo API for transparent compatibility with TraceQL clients.

Any Tempo/OpenTelemetry compatible client or application can be used with gigapipe out of the box

The Tempo datasource can be used to natively query traces including TraceQL and supporting service graphs

:tada: No plugins needed
:eye: No Grafana? No problem! Use View

TraceQL Metrics

gigapipe supports TraceQL metrics endpoints for calculating metrics from trace data. Calculate error rates, latency distributions, and throughput directly from traces using TraceQL queries.

Metrics endpoints:

/api/metrics/query_range - range queries over a time window
/api/metrics/query - instant queries at a specific time

Parameters:

q or query - TraceQL query string (required)
start - start timestamp (unix seconds, nanoseconds, or RFC3339)
end - end timestamp (same format as start)
since - relative time duration (e.g., "1h", "30m") - alternative to start/end
step - query resolution (e.g., "15s", "1m") - auto-calculated if not specified

Example:

GET /api/metrics/query_range?q={span.http.status_code=200}&since=1h&step=1m

This query calculates span counts for successful HTTP requests over the last hour with 1-minute resolution.

🔥 Pyroscope + Phlare

⚡ gigapipe implements the Pyroscope API for transparent compatibility with Pyroscope SDK clients.

Any Pyroscope SDK client or Pyroscope compatible agent can be used with gigapipe out of the box for continuous profiling

OTLP Profiles Ingestion

⚡ gigapipe also ingests the OpenTelemetry profiles signal (profiles/v1development) directly from OpenTelemetry collectors:

POST /v1development/profiles      Content-Type: application/x-protobuf

💡 No collector code changes required — point an otlphttp exporter at gigapipe and profiles flow in.

OTLP profiles land in the same ClickHouse profiles tables as Pyroscope data and are queryable through the existing Pyroscope-compatible API (flamegraph, series, stacktraces). See docs/otlp-profiles.md for details.

DOT Graph Rendering

The Render endpoint (/render?format=dot) generates Graphviz DOT format graphs for profiling visualization with enhanced readability:

Features:

maxNodes parameter: Limit the number of nodes displayed in DOT graphs (0 = unlimited). Use this to control graph complexity for large profiles.
```
GET /render?query=...&from=...&until=...&format=dot&maxNodes=50
```
Human-readable values: Automatically formats values based on unit type (e.g., "1.23s" for durations, "1.23 MB" for memory)
Variable font sizes: Node font size scales from 8pt to 24pt based on self-sample percentage, creating visual hierarchy where hotter code paths are more prominent