What Is a Cloud Native Database?

A cloud native database is a database that is designed to take full advantage of cloud technology and distributed systems.

Even though many databases can run in the cloud, there is a difference between cloud-ready and cloud native.

"Cloud native" describes services, software, APIs, or databases that are architected and built to be run and deployed on the cloud while benefiting from features that the cloud native systems offers.

Cloud native databases can run on cloud native technologies such as Kubernetes to deliver a flexible and scalable data storage and querying solution.

Cloud Native database and Kubernetes

When talking about cloud native, we must refer to containers because that is where cloud native applications run. Containers, however, are built and deployed to the cloud using technologies like Docker and Kubernetes. Therefore, to consider a database cloud native, it needs to run in containers and at the same time, it should be able to store data and ensure statefulness.

Persisting and moving data in the cloud is a major issue because Kubernetes was initially designed for stateless workloads. Because databases need to persist data, recent improvements have introduced StatefulSets and persistent volumes to Kubernetes, making it easy to run stateful workloads on Kubernetes.

Cloud native databases leverage these improvements to bring databases to the cloud to enjoy all benefits of Kubernetes, including elasticity and resilience.

Difference between cloud native and traditional database

With the increasing adoption of microservices and containerized applications, there is a need for a database that works similarly to the application to take its full advantage.

Traditional databases like MySQL and MongoDB are limited in many aspects, including scalability, security, and accessibility. Even though they can be integrated with the cloud, using these databases in the cloud limits the ability of applications to enjoy the benefits of cloud technology.

Below are some beneficial characteristics of cloud native databases that set them apart from traditional databases.

Advanced Scalability

Probably the most important characteristic of a cloud native database is its ability to scale with workloads. A cloud native database must be able to increase its capacity to accommodate an increase in workload dynamically.

This allows organizations to run their application without worrying about storage limits.

Elasticity

As important as it is to scale up, so it is to scale back down. A cloud native database must be able to scale down or reduce its capacity when the workload reduces to ensure that you pay for only the resources you need - which is one of the benefits of the cloud.

Resiliency

A cloud native database must be able to survive system failures without losing any data. When there is a failure in a part of the system, cloud native databases can move data to a new pod and heal automatically.

Automation

Running a database in Kubernetes allows us to codify the process of deploying and managing cloud native databases to achieve automation.

Accessibility

Unlike traditional databases, a cloud native database that can only be accessed through the deployment system, cloud native uses distributed database technology, making it easily accessible.

Reduced Cost

One of the dominant characteristics of the cloud is the ability to pay for only resources that you use. Coupled with being elastic, a cloud native database allows you to pay as you go and pay for only resources that you need.

Easy Database Management

The cloud native database makes it easy to manage and update the database due to its characteristics like automation and scalability.

Examples of cloud native database

Spanner

Spanner is a cloud native relational database developed by Google. It offers various cloud native features, including scalability, consistent reads, and resiliency through multi-site replication. It is a scalable solution used by major companies, including Uber and Lyft.

CockroachDB

CockroachDB is an open source relational database technology that offers high resilience, scalability, and data distribution. It is a cloud native database solution ideal for various use cases, including customer relationship management, financial transactions, and record systems.

K8ssandra

K8ssandra is a NoSQL Kubernetes-native database solution built on the Apache Cassandra database. It provides a distributed, scalable, highly available, and production-ready cloud native database solution that runs on Kubernetes clusters.

Vitess

Vitess is an open source cloud native database built on traditional MySQL. Vitess makes clustering and horizontal scaling of MySQL databases easy.

Vitess uses the Kubernetes operator pattern to run MySQL database operations on Kubernetes.

It is a scalable, extensible, and highly efficient cloud native database solution used by major companies such as Hubspot and Slack.

How slack uses Vitess

Using the traditional MySQL database, Slack had major issues with availability, hitting the host's scale limits, and being stuck with one data architecture model.

Slack’s database grew bigger and introduced new features such as Slack Connect, which requires a more flexible and scalable database solution.

To solve this problem, Slack sought after Vitess because it is compatible with MySQL (which was their native database); it is scalable, operable, and extensible - which matches all requirements of their applications.

The engineering team at slack then built a prototype for migrating their data from the traditional database to Vitess, and they recently announced that they had moved 99% of their database to Vitess - helping them solve their database problems.