# Parseable

Parseable is a columnar data lake, purpose built for observability. 

Built in Rust on Apache Parquet and object storage (S3, GCS, Azure Blob, MinIO), it delivers up to 90% data compression, PostgreSQL-compatible SQL queries, native PromQL support (Enterprise), OpenTelemetry ingestion, and AI-native features. 

Parseable is available as a single binary or containerized image on any cloud or on-premises, with no external dependencies beyond an object store.

It is particularly well-suited for matured teams that are looking to keep observability within their own infrastructure. Parseable is designed to be self-contained, highly efficient, and easy to deploy, ensuring that teams can maintain full control over their observability data without relying on external services.

Parseable is available in three tiers: **open source** (AGPLv3, self-hosted, free), **Cloud** ($0.39/GB, fully managed on Parseable cloud, 14-day free trial), and **Enterprise** (dedicated infrastructure, BYOC).

## Key Differentiators

- Open, Columnar format
    - Compression of up to 90% on observability data.
    - Data is always accessible, even outside of Parseable allowing complete control and flexibility for teams.
    - Moves indexing and query processing closer to the data, improving performance and reducing latency.

- Object store first design
    - Clean segregation of storage and compute layers, allowing independent scaling and management.
    - Simpler architecture for day 2 operations. Much simpler to run at scale compared to traditional monolithic observability solutions.

- Single binary to centralize telemetry
    - Zero external dependencies required.
    - In contrast to running so called stacks of separate observability tools, Parseable centralizes all telemetry into a single binary.

## Getting started

- [Quickstart](https://www.parseable.com/docs/get-started)
- [Architecture](https://www.parseable.com/docs/architecture)
- [Self-Hosted Installation](https://www.parseable.com/docs/self-hosted/installation): Kubernetes (Helm), Docker Compose, or bare metal. Distributed mode requires an object store.
- [Ingestion overview](https://www.parseable.com/docs/ingestion): HTTP(S) JSON or OTLP/HTTP. Required headers: `X-P-Stream` (dataset), `Authorization`. Max payload 10 MiB. Events staged to disk, flushed to Parquet every minute.
- [OpenTelemetry](https://www.parseable.com/docs/ingest-data/otel): Native OTLP/HTTP for logs, metrics, and traces — zero configuration.
- [SQL / Key Concepts](https://www.parseable.com/docs/key-concepts/query): PostgreSQL-compatible SQL via Apache DataFusion. Supports aggregate, window, and scalar functions, regex operators (`~`, `~*`, `!~`), `EXPLAIN ANALYZE`, and `?fields=true` for response metadata. Every query requires a `start`/`end` timestamp.
- [Integrations](https://www.parseable.com/docs/integrations)
- [Pricing](https://www.parseable.com/pricing)

## Optional: full documentation

- [llms-full.txt](https://www.parseable.com/llms-full.txt): Complete single-file documentation for deep LLM analysis and summarisation.
- [GitHub](https://github.com/parseablehq/parseable): Source code, issues, and contribution guide.
- [Docs home](https://www.parseable.com/docs): Interactive documentation site.
- [Live demo](https://app.parseable.com): Try Parseable without installing anything.