Overview

The Data Streamhouse supports a variety of deployment setups tailored to different environments and organizational needs. Whether you're deploying for a proof of concept, a single project, or as part of a fully integrated enterprise platform—this section will help you identify the right starting point.

This guide does not cover the free version of Data Streamhouse.

Choosing the Right Setup

Before deploying, clearly define your deployment strategy and use cases. The structure of your environment—particularly network topology—will determine the optimal deployment model.

Ask yourself: Is one or more Kafka clusters located in an isolated network that cannot be accessed externally?

If no, a centralized setup with a single Data Streamhouse instance is recommended. This enables full feature access, including cross-environment insights & management, migration tooling, and real-world test data generation across dev/stage/prod environments.

If yes, isolated environments (e.g., air-gapped production) will require a separate, local Data Streamhouse deployment (see Basic Setup) within that network.

You’ll find a list of common scenarios at the end of this section. Each is described in detail on its respective page.

Identity Providers & Right Management

The Data Streamhouse integrates with enterprise identity providers via LDAP or OpenID Connect—including platforms such as Okta, Entra ID, or Active Directory. When enabled, user authentication is delegated to these providers.

You can manage user groups and roles manually within the Data Streamhouse, even when using external authentication. However, we recommend enabling automatic group mapping, which aligns Data Streamhouse groups with existing assignments in your identity provider—simplifying and centralizing access control.

For PoCs or small-scale environments, internal authentication and authorization mechanisms can be used to get started quickly without identity provider integration.

See the Identity Provider Integration section for step-by-step instructions to configure and enable LDAP/OpenID authentication.

License Activation

The Data Streamhouse is often deployed in high-security environments without internet access. In such cases, license activation must be performed manually, as the system cannot reach the Xeotek Link license server.

The License Activation Scenarios section provides detailed guidance for both online and offline activation flows.

Common Deployment Scenarios

1. Basic Setup

Ideal for:

• Quick start / PoCs

• ≤5 Kafka clusters

• Single non-isolated network

Key factors:

• Direct network access to all connected Kafka clusters

• Minimal infrastructure footprint

Components:

• 1× Portal

• 1× Argus

• 1+ optional Machina instances (analytics, alerts, etc.)

2. Advanced Setup

Ideal for:

• Large-scale data streaming environments

• >5 Kafka clusters

• 100k+ data streams

• Hundreds of users

• No isolated networks

Key factors:

• Central Portal connecting to all streaming environments

• Clustered, high-availability configuration

• Horizontally scalable

Components:

• 1× Portal Cluster (3+ instances for HA)

• 1× Argus per 4 Kafka clusters

• 10+ optional Machina instances (depending on analytics workload)

3. Automated Deployment

Ideal for:

• IaC-driven environments

• ≤5 Kafka clusters

• Single isolated network

• Lightweight setups (no computation layer)

Key factors:

• Fully automatable via Terraform, Ansible, etc.

• No manual interaction during deployment

• Focused on monitoring and exploration

Components:

• 1× Portal

• 1× Argus