CockroachDB: The Definitive Guide

Chapter 9. Planning a Deployment

In the preceding chapters, we’ve described how to get started with CockroachDB and how to develop highly available and performant applications with the CockroachDB platform.

Now it’s time to consider how to set up a production CockroachDB cluster to support your application.

The distributed nature of CockroachDB allows for a large range of deployment topologies. Choosing the right topology requires an understanding of your application’s requirements and the cost and performance implications of various CockroachDB options. In this chapter, we review the steps in planning a deployment and provide an overview of the most common deployment patterns.

There are two main categories of CockroachDB deployments:

• A fully managed CockroachDB Cloud deployment, in which all aspects of CockroachDB cluster management are handled by the CockroachDB dedicated cloud platform. Within CockroachDB Cloud, you can choose from serverless or dedicated hardware options.

• A self-hosted or “do it yourself” deployment in which you install CockroachDB on your own hardware platform or on cloud-based virtual machines (VMs) that you have available.

In any of these scenarios, you may deploy single-region or multiregion topologies.

Within a self-hosted deployment, you have two additional dimensions of choice. Whether using on-premise hardware or cloud resources, you can install CockroachDB directly onto the OS, or you can install CockroachDB into a Kubernetes cluster.

Know ...