---
title: How to create a Cluster for Apache Kafka®
description: Learn how to create a Cluster for Apache Kafka® with Scaleway, and get your hosted, fully-managed Kafka infrastructure up and running with just a few clicks.
dates:
  validation: 2026-06-19
  posted: 2025-09-15
---
import Requirements from '@macros/iam/requirements.mdx'
import VPCRoutingCompatibility from '@macros/vpc/vpc-routing-compatibility.mdx'

This page shows you how to create a Cluster for Apache Kafka® via the Scaleway console.

<Requirements />

  - A Scaleway account logged into the [console](https://console.scaleway.com)
  - [Owner](/iam/concepts/#owner) status or [IAM permissions](/iam/concepts/#permission) allowing you to perform actions in the intended Organization

## How to create a Cluster for Apache Kafka®

1. Click **Apache Kafka®** in the **Data and Analytics** section of the [console](https://console.scaleway.com/) side menu. The **Clusters for Apache Kafka®a** product page displays.

2. Click **Create cluster**. The creation wizard displays.

3. Select a **region** for the cluster.

4. Select a **Kafka® version** to run on the cluster.

5. Select an Availability Zone (AZ) configuration for the cluster:
    - **Multi-AZ**: Select this configuration to create a highly available Apache Kafka® cluster. Data can be partitioned and replicated across three nodes (brokers) for maximum fault tolerance and availability. Currently, the multi-AZ offers replicate three nodes over two Availability Zones.
    - **Single-AZ**: Select this configuration to provision an Apache Kafka® cluster in a single AZ without compromising on fault-tolerance and replication.

6. Select a **node type**. Choose between [dedicated or shared nodes](/clusters-for-kafka/concepts/#node), depending on your workload requirements and use case. For shared nodes, various types are available, each offering different cost-performance ratios.

7. Select the **number of nodes** (brokers) you want in the cluster.
    - Allowed number of nodes: 1 - 30, in multiples of 3 (except for a single‑node option).  
    - For highly‑available clusters enabling data replication: use at least 3 nodes.  
    - For testing and development: 1 node is sufficient.

    <Message type="note">
    A single‑node cluster is not compatible with a multi‑AZ configuration.
    </Message>

    <Message type="note">
    The number of nodes cannot be changed after cluster creation.
    </Message>

8. Choose the **storage capacity** to define how much Block Storage to provision for each node.

9. Configure **network connectivity**. Currently, connectivity to clusters is facilitated exclusively through **Private Networks**. Select a Private Network from the dropdown list to attach your cluster to.

    <VPCRoutingCompatibility />

10. Create **credentials** by defining a username and password. These will be used when connecting to your cluster. Currently, only one user can be created per cluster, and the username is not modifiable after creation (though the password can be changed).

11. Enter a **name**, and (optionally), **tags** to identify your cluster.

12. Check the estimated cost summary, then click **Create cluster**.

Your cluster is created, and you are taken to its **Overview** page in the console.