Deployment Overview

Kafka Mobility Intelligence (KMI) is a tool for migrating Kafka workloads to Confluent. This section covers how to install and run the suite in your own environment.

KMI is distributed as a self-contained package that includes all container images and configuration files needed for deployment. There are two supported installation methods.

Choose a Deployment Method

Air-Gapped (Docker Compose)

Single-node deployment using Docker Compose, designed for offline environments without internet access. All images ship inside the package. See Air-Gapped Deployment.

Kubernetes (Helm)

Multi-platform deployment using Helm charts across AWS EKS, GCP GKE, Azure AKS, OpenShift, and on-premises clusters. Supports private registries, cloud secret managers, and zero-downtime rolling updates. See Kubernetes Prerequisites.

Which Method Should I Use?

Consideration	Air-Gapped (Docker Compose)	Kubernetes (Helm)
Environment	Single host, offline	Kubernetes cluster
High availability	Not supported	Supported via replicas
Secret management	.env file	.env or cloud secret manager
In-place updates	Not supported (redeploy)	Rolling updates with zero downtime
Best for	Isolated or evaluation installs	Production and multi-platform installs

Note

Both methods ship all container images inside the package, so neither requires the target environment to pull images from the public internet.

Common Concepts

Regardless of the method you choose, deployment involves the same core steps:

1. Transfer and extract the deployment package on the target environment.
2. Create a configuration file from the provided example template.
3. Generate application secrets (JWT_SECRET and AES_KEY).
4. Run the deployment script.
5. Verify that all services are running and access the web interface.

Caution

KMI requires a valid license key for operation. After the first sign-in, the system prompts for a license key before granting access to the dashboard. Contact your administrator to obtain a license.