Hadron

Paid Self-hosted

What Hadron is

Hadron is a self-hosted distributed data storage system for ingesting events and organizing work into multi-stage structured workflows. Its Streams keep events in append-only logs: records are immutable, each partition has an absolute order, and Streams can scale horizontally. Pipelines coordinate stages using Stream data and can begin when an event matches a configured trigger. For other communication, Exchanges handle ephemeral messages and Endpoints provide general-purpose RPC handlers. Hadron uses the CloudEvents 1.0 model for Stream data, Pipeline inputs and outputs, and ephemeral messages.

Hadron is built to run in Kubernetes and is installed with Helm and Kubernetes configuration resources. Streams can remain available as long as at least one partition is running. Retention can be set to a time policy, defaulting to seven days, or to retain data indefinitely. Producers and Consumers can be written in any language; the project maintains a shared Rust client and command-line interface. Components expose Prometheus metrics over HTTP, with integration options for Prometheus scraping and the OTEL collector. Access uses Tokens that generate JWTs signed by the Operator’s private RSA key, with permissions set by resource. The CLI documentation says external cluster access is in progress. Hadron Cloud is described as under construction.

Who it suits

Hadron suits teams building event based workflows that can deploy and manage software in Kubernetes. It may also suit teams that want configurable Stream retention, language-flexible clients, and HTTP metrics for monitoring.

What it does well

Streams are append-only and scale horizontally.
Pipelines can trigger on matching event types.
Stream data can be retained indefinitely.
Producers and Consumers can use any language.
Components expose Prometheus metrics over HTTP.

What to know first

Deployment requires a Kubernetes environment and Helm.
CLI access from outside the cluster is in progress.
Hadron Cloud is still under construction.

Limits

Stream retention defaults to seven days unless configured with another time policy or indefinite retention. Streams remain available when at least one partition is running.

Getting started

Visit the Hadron website.
Prepare a Kubernetes deployment.
Install Hadron using Helm and Kubernetes configuration resources.
Choose Stream retention and configure resources and permissions.

Verdict

Choose Hadron if you need self-hosted event storage with ordered, immutable Streams and multi-stage Pipelines in Kubernetes. Consider the deployment and access details carefully if your team needs external CLI access or a hosted managed service now.

Questions

Is Hadron free?

Yes. Hadron is free and open source.

Where does Hadron run?

Hadron is self-hosted and designed for Kubernetes. It is installed using Helm and Kubernetes configuration resources.

Which client languages are supported?

Producers and Consumers can be written in any language. The project maintains a shared Rust client and a CLI.

How does Hadron handle event data?

Streams are append-only, immutable logs with ordering per partition. They use the CloudEvents 1.0 model, as do Pipeline inputs and outputs and ephemeral messages.

Can I access Hadron from outside Kubernetes?

The CLI documentation says access from outside the Kubernetes cluster is in progress.

Is a hosted service available?

Hadron Cloud is described as under construction as the hosted, fully managed option.

Facts from Hadron's own pages

Access control: Hadron authenticates with Tokens that produce JWTs signed with the Operator's private RSA key, with permissions configurable by resource. · hadron-project.github.io, 4 Oct 2026
Authentication: Hadron authentication uses Tokens that the Operator processes into JWTs stored in Kubernetes Secrets. · hadron-project.github.io, 11 Oct 2026
CLI access limit: The CLI documentation says access from outside the Kubernetes cluster is in progress. · hadron-project.github.io, 11 Oct 2026
Client library: The official clients page lists a Rust client and says client libraries for other languages are planned as Hadron matures. · hadron-project.github.io, 11 Oct 2026
Clients: Producers and Consumers can be written in any language, and the project maintains a shared Rust client and a CLI. · hadron-project.github.io, 4 Oct 2026
Event format: Hadron uses CloudEvents 1.0 for data entering and leaving the system. · hadron-project.github.io, 11 Oct 2026
External access: The CLI documentation says access from outside the Kubernetes cluster is in progress. · hadron-project.github.io, 4 Oct 2026
Hosted service: The quick start describes Hadron Cloud as under construction and the hosted, fully managed option. · hadron-project.github.io, 4 Oct 2026
Hosted service status: The quick start describes Hadron Cloud as under construction and hosted by the Hadron project team. · hadron-project.github.io, 11 Oct 2026
Installation: The quick start installs the Hadron Operator with Helm and requires Helm version 3.7.0 or later for OCI usage. · hadron-project.github.io, 11 Oct 2026
Intended users: The guide presents Hadron for distributed applications with many microservices and event or telemetry signals, and includes a local development use case. · hadron-project.github.io, 11 Oct 2026
Kubernetes: Hadron is designed to run in Kubernetes, and its Operator manages cluster lifecycle tasks including upgrades, horizontal scaling, and access control. · hadron-project.github.io, 11 Oct 2026
Messaging and RPC: Hadron describes Exchanges as ephemeral messaging between processes and Endpoints as general-purpose RPC handlers. · hadron-project.github.io, 11 Oct 2026
Monitoring: Hadron components expose Prometheus metrics over HTTP, with documented integration options including Prometheus scraping and the OTEL collector. · hadron-project.github.io, 4 Oct 2026
Pipelines: Pipelines orchestrate multi-stage workflows for Stream data, triggered when an event type matches a Pipeline trigger. · hadron-project.github.io, 11 Oct 2026
Product: Hadron is a distributed data storage system for ingesting events and working with them in multi-stage structured workflows. · hadron-project.github.io, 4 Oct 2026
Purpose: Hadron is a distributed event storage and messaging system for ingesting events and working with them in multi-stage workflows. · hadron-project.github.io, 11 Oct 2026
Retention: Stream data retention supports a configurable time policy that defaults to seven days, or an indefinite retain policy. · hadron-project.github.io, 4 Oct 2026
Scaling and availability: The documentation says Streams can scale horizontally and remain available when at least one partition is running. · hadron-project.github.io, 4 Oct 2026
Streams: Streams are append-only immutable logs with absolute ordering per partition and configurable retention, which defaults to seven days. · hadron-project.github.io, 11 Oct 2026
TLS setup: The quick start says cert-manager is optional and simplifies TLS certificate setup for Hadron's validating webhooks. · hadron-project.github.io, 4 Oct 2026

Where Hadron stands on Freedom251

Event Streaming Platforms#19 of 27