MariaDB Cloud: A Semi-Technical Introduction

Written by Written by Kaj Arnö Leave a comment on MariaDB Cloud: A Semi-Technical Introduction
MariaDB Cloud

When co-founding MariaDB plc under the name SkySQL Corporation fifteen years ago (before MariaDB Foundation existed), we picked the name SkySQL with a bit of a reference to the cloud. Obviously, we didn’t then have a clear understanding of what type of a cloud offering MariaDB should have. I am glad to note that those days are finally and completely over, with MariaDB Cloud. I wish it hadn’t taken fifteen years!

So now, I sat down for a proper chat with Jags Ramnarayan, the technical father of SkySQL / MariaDB Cloud. I wanted to properly understand what we should ideally have understood in 2009.

Our conversation explored the goals, the technology, and the outcome: Firstly, what makes MariaDB Cloud distinct? Secondly, how does its architecture and goals extend the MariaDB tradition into a modern, cloud-native era?

The result? Rest assured, I am very happy about MariaDB Cloud – also in my role as Executive Chairman of MariaDB Foundation.

So, buckle up for a structured narrative — one that builds understanding step by step:

  • We begin with the concept of “Serverless”, both in general and in the specific context of MariaDB.
  • Next, we look at how MariaDB Cloud relates to its Community and Enterprise counterparts, and at the technical goals that shape its evolution.
  • Once the objectives are clear, we turn to the technologies and AI integrations that make them possible.
  • Finally, we look outward — at the clouds currently supporting MariaDB Cloud, and at how its design ensures true portability across providers.

1. What Does “Serverless” Mean – Traditionally and for MariaDB?

In most computing literature, serverless describes systems that automatically handle infrastructure provisioning and scaling — CPU, memory, and storage — so that developers can focus purely on code or data.
The cloud platform takes care of allocating resources on demand, invisibly.

Of course, Serverless doesn’t literally mean “MariaDB Server without a server”. In the MariaDB context, the concept means:

  • Users don’t manually configure or manage database servers.
  • Provisioning and scaling happen automatically and instantly.
  • The experience feels as seamless as working locally — yet runs on elastic cloud resources.

In traditional setups, database administrators used to tweak configuration files such as my.cnf to fine-tune memory, cache, replication, and connections. In MariaDB Cloud, this manual tuning is replaced by automated configuration and scaling logic — a system that optimises dynamically what my.cnf would have done manually.

The result: a self-managing MariaDB Server that can scale up or down within milliseconds, while preserving cached data and stable performance. As someone having tweaked a few my.cnf files in my life, I see that as a worthy outcome!

2. MariaDB Cloud, MariaDB Community Server, and MariaDB Enterprise Server

Currently, the serverless product in MariaDB Cloud runs on the MariaDB Community Server, while MariaDB Enterprise Server will soon be introduced in the provisioned version.

From a user’s point of view, MariaDB Cloud is meant to behave exactly like a self-managed MariaDB instance — except that all provisioning, scaling, and configuration (including what once lived inside my.cnf) is automated by the platform.

Beyond that, MariaDB Cloud goes further: it handles security (both on the wire and at rest), automates backups, and includes built-in monitoring for a fully managed experience.

A free developer version is available, limited in compute and storage, but perfect for experimentation, testing, and lightweight workloads.

3. Key Technical Goals of MariaDB Cloud

Now we come to the core of what MariaDB Cloud is meant to be, according to Jags: MariaDB Cloud combines open-source reliability with cloud-native automation. Its technical goals include:

  1. Instant elasticity: resources scale within milliseconds.
  2. Operational transparency: caching and I/O remain stable even during dynamic scaling.
  3. Ease of onboarding: new databases launch within seconds.
  4. Fault tolerance: even a single-server deployment can recover instantly after a crash, with automation resurrecting the server and preserving app connections.
  5. Pay-per-use efficiency: customers pay only for what they actually consume — not for pre-provisioned capacity.
  6. Automatic configuration: replaces manual editing of my.cnf with self-managing tuning.
  7. Developer empowerment: AI-assisted tools guide data loading, transformation, and SQL creation.

In short, MariaDB Cloud is designed to deliver the MariaDB experience as a resilient, elastic, and cost-efficient managed platform — powered by intelligent automation.

4. Development Languages and Tools Used by MariaDB Cloud

Now, given the “semi technical” goal of mine, I’m not happy until I have a basic understanding of the underlying architecture. What technologies does MariaDB Cloud build on? In which language(s) was it coded?

The answer: MariaDB Cloud relies on a wide ecosystem of modern technologies designed for automation, security, and multi-cloud operation:

  1. Kubernetes – orchestrates containerised MariaDB instances.
  2. HashiCorp Terraform – abstracts cloud APIs and manages infrastructure as code.
  3. Docker – packages MariaDB environments for reproducibility and rapid deployment.
  4. Additional ecosystem components – including an external billing provider, authentication service, and integrated support system, plus several Kubernetes CNCF projects and AI-related frameworks such as LlamaIndex and OpenAI.

The stack is primarily written in Go, Python, and Terraform DSL, combining speed, portability, and a strong ecosystem.
Together, these technologies enable MariaDB Cloud to deliver consistent, secure, and intelligent database environments across clouds with minimal human intervention.

5. The Relationship Between MariaDB Cloud and AI

I remember seeing demos of AI integration with MariaDB Cloud, and as someone passionate about the opportunities offered by combining MariaDB with AI, I was keen to explore the relationship between MariaDB Cloud and AI.

A defining aspect of MariaDB Cloud is its integration of AI assistants directly into the user interface, Jags explains:

  • Developers can generate SQL or table structures from plain-text or JSON.
  • Built-in AI agents interpret nested data and produce optimal schemas automatically.
  • Multiple large language models (LLMs) are supported to analyse and optimise SQL workloads.
  • The platform also supports custom AI database agents, built using the UI assistant and deployed directly in-app. These agents enable natural-language conversations over application datasets via a simple REST API, embedding conversational AI at the data layer.

All of this serves the goal of making data interaction and transformation “really straightforward and simple for developers”, allowing them to focus on application logic rather than syntax.

6. Clouds Currently Supported by MariaDB Cloud

So what bigger cloud is MariaDB Cloud part of?

Today, MariaDB Cloud runs on the world’s leading public clouds:

  • Amazon Web Services (AWS)
  • Google Cloud Platform (GCP)
  • Microsoft Azure

Support for additional providers is already planned — including secondary and sovereign clouds, particularly in Europe — ensuring compliance with regional data-residency and sovereignty requirements.

7. Portability of MariaDB Cloud Across Clouds

So what about cloud interoperability? And the technical hurdles in supporting new clouds?

MariaDB Cloud’s architecture enables true multi-cloud portability, using Terraform and Kubernetes to abstract away vendor-specific APIs.
This makes deployments on new environments fast — but each new provider still requires:

  • Rigorous testing and validation,
  • Security hardening, and
  • Ongoing baseline operational costs.

As Jags Ramnarayan noted, the decision to add a new cloud is therefore both a technical and a business decision — one balancing readiness with real customer demand.

Closing Thoughts

So what’s the outcome? Mind you, this is not a kicking-the-tyres type of blog entry. That will follow soon. It’s about understanding the main concepts and goals.

As I see it, MariaDB Cloud represents the logical continuation of the MariaDB mission — combining the openness of community software with the automation, resilience, and intelligence of modern cloud computing.

If our tests show that it delivers on its vision and promise, the outcome is good: For developers, it offers immediacy and simplicity; for enterprises, it delivers reliability and portability.

For the MariaDB ecosystem, this marks a new phase in which innovation and open collaboration continue, now in the cloud.