# Installation

We built Kloudfuse to run in a Kubernetes cluster on GCP, AWS, or Azure cloud.

## Prerequisites

To install the Kloudfuse stack, you must have the following:

**Basic Configuration**  
- Machine with at least 16vCPU, and 128GB memory.  
- A Kubernetes cluster; can be separate or shared with the application  
- Persistent volumes; Installation creates these; the installation account must have the necessary permissions  
- Helm version 3.10.0 or later  
- Kubectl

For advanced and production-grade customizations, see the relevant prerequisites. See various migration scenarios and integrations to select installation options that match your business requirements.

## Installation steps

- [Install Kloudfuse Using Helm](https://docs.kloudfuse.com/platform/3.5.0/install/)
- [Configure cloud-specific helm values](https://docs.kloudfuse.com/platform/3.5.0/install-ingress/)
- [Implement Multi-Zone Availability](https://docs.kloudfuse.com/platform/3.5.0/install-multi-zone/)
- [Add Users](https://docs.kloudfuse.com/platform/3.5.0/install-add-users/)
- [Access Kloudfuse UI](https://docs.kloudfuse.com/platform/3.5.0/install-ui/)
- [Configure Authentication](https://docs.kloudfuse.com/platform/3.5.0/sso/)
- [Configure the Object Store for Pinot](https://docs.kloudfuse.com/platform/3.5.0/install-object-store/)
- [Pinot Stream Isolation](https://docs.kloudfuse.com/platform/3.5.0/install-stream-isolation/)
- [Uninstall Kloudfuse](https://docs.kloudfuse.com/platform/3.5.0/uninstall/)
