Data is the heart of every modern application today. However, running databases reliably requires much more attention and knowledge than just installing the software.

The times when companies had teams of database administrators (DBAs) are long gone. Some organizations still maintain these teams, but this is mainly the case for large, regulated organizations where databases are critical, such as financial or medical institutions.

In most organizations, all operations related to databases – updates, backups, monitoring, security, scaling, high availability, and what requires even more expertise – design, performance, and optimization are in the developer teams’ responsibilities. All of these activities require the team’s time and focus, not to mention specialized knowledge.

This effort is greatly relieved by switching to managed databases. However, not everything is taken off the team’s shoulders. In this article, we discuss what a managed database is, why we should consider adopting one in our organization, and the advantages and disadvantages of this solution.

What is a managed database?

A managed database service is commonly referred to as “Database-as-a-Service” or DBaaS. It is a database operated by a service provider and offered on a subscription basis. This means that instead of provisioning servers, installing databases, and patching and updating them, users can create databases ready for application use.

The provider is responsible for all operational elements, such as:

• Provisioning infrastructure
• Installing the database and all client software on the provisioned infrastructure
• Maintaining the instance and database security, operational patches, and updates
• Managing failures for the infrastructure
• Scaling operations
• Replicating on the operational level
• Backups (in the form of snapshots)

The user is responsible for:

• Designing the database schema
• Implementing schemas, queries, etc.
• Managing database users
• Access management
• Ensuring performance and consistency of data
• Backups (in the form of database dumps)

The provider is responsible for provisioning and operating the database platform, while your team needs to handle the data, access permissions, and how the applications use the database.

Why should organizations consider utilizing managed databases?

The biggest advantage of a managed database is the fact that DBaaS reduces operational complexity. Routine tasks such as software updates, backups, monitoring, or infrastructure maintenance are handled by vendors. It means that the team’s workload is significantly reduced, and the required knowledge consists of database operations on the application level.

Managed Databases also improve reliability. Automated backups, built-in monitoring, or high availability are typically available out of the box, without any additional configurations.

The final reason concerns time-to-market. Managed databases eliminate the time required to provision infrastructure, install databases, and configure them. It simplifies and shortens the time from idea to deployment to a minimum.

What are the advantages and disadvantages of managed databases?

Every solution must be chosen to serve a proper purpose. Which approach to select is part of architectural decisions, mainly those we call Quality Attributes (formerly known as Non-Functional Requirements). Business Requirements will tell us what the business needs for a database, and Quality Attributes describe all aspects that are operational, functional, managerial, and more.

However, there is one consideration that is part of business requirements. The costs.

Let’s go through a few advantages and disadvantages of Managed Databases to help us better understand which approach we should choose.

Advantages of managed databases

• Reduced operational overhead. This is one of the biggest advantages. The organization doesn’t need to hire skilled engineers to Manage Databases. Developers can quickly request an instance and start working without waiting for provisioning, configuration, or other steps. This increases the team’s productivity and reduces operational overhead.
• Automated backups. Operations are managed and overseen by the provider. It ensures that copies are made, stored securely and reliably, and are easy to restore. Additionally, using these snapshots, users can create additional database instances to run tests or create new environments without additional operations.
• Faster deployments. Both points above lead to a much shorter time-to-market for the features. Provisioning takes minutes, and copies are automated, these allow development teams to work much faster.
• Higher reliability. In general, all managed database services have built-in high availability. The implementation of the solution might differ between providers, but in the end, we receive a service that is reliable, heals quickly, and where any issue with the underlying infrastructure shouldn’t cause anything more than a short connection blip. Of course, according to Murphy’s law, if something can go wrong, it will go wrong, so we cannot forget that there is actual infrastructure and real humans who deliver the service, so more serious issues can happen. But these issues are less likely to occur than in services we operate ourselves entirely.
• Automated and simplified scaling. For us as users, scaling is extremely easy. We just say that we want scaling, how many replicas to use, and so on. Everything is configured behind the hood without us knowing how. This makes the whole operational part almost non-existent.
• Improved security. Although we need to remember, it is not just the vendor’s responsibility. According to the shared responsibility model, there are parts where we still need to be careful and understand how the service is configured. To list a few elements we still need to take care of:
  • How the networking is configured, what ports and networks are able to connect
  • How we encrypt the data, how and where we store the keys, whether we rotate them, and whether we own these keys.
  • Who has access to the database and the data stored
  • Who has access to the service itself
• Predictable operations. With managed services, we rely on the vendor’s knowledge. Our vendor is probably more skillful and experienced than we are in this specific field. They created the service, which requires a deep understanding of the software they run. And they run hundreds or thousands of service instances for their customers. They understand how the service behaves, what needs to be done, and how to operate it.
• Business savings. No or minimal costs for skilled engineers, together with much faster deployment times, can lead to significant savings.

Disadvantages of managed databases

Unfortunately, nothing has only a bright side. With all the advantages we just discussed, DBaaS comes with its own considerations that may be risky or at least inconvenient for our business.

• Less infrastructure control. We don’t own the machine. Even more, we don’t have access to it. We don’t understand its behavior, actions taken, or performance. We rely only on information provided by vendors, like some logs and notifications.
• Potential vendor-lock. Without the proper skill set, we cannot easily change the vendor (if at all). We have a closed machine, a black box without access. Backups are done by the vendor in their own way. We don’t have anything that we can use elsewhere. Of course, the database is still a database. For example, if we dump data from a PostgreSQL run on vendor A’s infrastructure, we can copy the dump to vendor B’s infrastructure and restore it there. But this requires elementary knowledge of system and database operations, which is not necessarily part of developers’ skill sets.
• Higher operational costs. Managed services are more expensive than services we provision and manage ourselves. This well-known fact stems from the required work and the perks it provides. Someone designed the service and is further developing it. It must be maintained and controlled. Depending on the vendor and the service, the difference can be really significant.
• Less flexibility. We receive some prepared “box” with the service. Sometimes we have some room to configure parameters, but in most cases we have to use what we get. Vendors cannot prepare service for every single case. It is simply impossible. That is why they select the most common scenarios and prepare the service for these scenarios. If we need, for example, extreme performance, we need to configure the machine, software, and application ourselves.

When is a managed database the right choice?

For most organizations building cloud-based applications, a managed database should be a preferred option. It is a valuable choice for teams that want to reduce operational overhead, accelerate delivery, and improve service reliability without investing in dedicated engineers.

Organizations with specialized requirements, regulatory constraints, or a need for complete control can still manage the database themselves. However, most modern cloud-native applications will benefit from managed services.

Does UpCloud offer managed databases?

UpCloud offers managed databases and provides 4 different engines:

Relational database

• MySQL
• PostgreSQL

Non-relational database

• OpenSearch
• Valkey (fork of Redis)

All these databases are available for all users and offer features as follows:

• Build your service with 1 node for development, and up to 3 nodes in the cluster to achieve high availability. For OpenSearch, the maximum is 15 nodes.
• Backups. We can enable automated backups for each database and plan
• Scaling without downtime
• Uptime with 99.999% SLA
• Enhanced security and network configurations
• No fees for data transfer

To start working with UpCloud’s DBaaS, simply log in to the console, navigate to Managed databases, select one type, and start deploying!

Conclusion

Managed Databases are much more than databases cared for by someone else. It is a service that removes much of the operational burden of running critical parts of applications. By outsourcing routine maintenance and management to a provider, organizations can reduce complexity and cognitive load, improve reliability, and enable engineers to focus on delivering business value.

Check our Managed Databases for more information about each engine and the options we offer.

If you would like to learn more about the service, here is the documentation for MySQL, PostgreSQL, OpenSearch, or Valkey, and also check our tutorials and videos!

If you have any further questions, don’t hesitate to ask our team, we are here for you!