Open source vs. proprietary database management

High demand for both open source and proprietary DBMS software

Demand for both open source and proprietary database systems remains high and is growing strongly. In a report on cloud databases published in December 2023, Gartner said the worldwide DBMS market grew 14.4% to $91 billion in 2022 and was likely to top the $100 billion mark for 2023 once the final revenue numbers are in. Meanwhile, the top technologies in the DBMS popularity rankings on the DB-Engines website include a mix of proprietary and open source databases.

"Everyone needs these systems," observed William McKnight, president of McKnight Consulting Group. Organizations, he said, must have clean, reliable data that is secure and compliant with all applicable regulations so executives and other employees can use it to glean insights, drive business results and -- increasingly -- power intelligent systems.

All of this necessitates good overall data management supported by various tools, including database management systems, McKnight added.

It also requires data management leaders to diligently consider whether an open source DBMS or a proprietary one will be the better choice to help them deliver on those points.

Proprietary database management software pros and cons

Proprietary software's source code is unavailable to users or other vendors; as a result, it's also referred to as closed source software. It's sold by vendors as off-the-shelf commercial products that might or might not be customizable by users. Although that might sound overly rigid, IT advisors stressed that proprietary DBMSes offer multiple benefits to organizations.

First, proprietary database vendors deliver products with a full range of complementary capabilities, said Noel Yuhanna, vice president and principal analyst at Forrester Research.

For example, these vendors build in integrations so enterprise teams can more quickly and easily deploy databases -- an important point for organizations with large and complex technology stacks. They have also added more automation to DBMS software in recent years and increased amounts of intelligence. And these systems generally have advanced security capabilities, as vendors have recognized the high value that organizations place on securing their data.

In addition, proprietary DBMSes tend to be highly optimized, finely tuned and well tested.

As a result of all this work that vendors put into the systems they sell to organizations, Yuhanna said proprietary databases are highly reliable and "can deliver high performance at scale [with] security, innovation and automation."

Organizations also get vendor support when opting for proprietary database management technologies, and they typically find that it's easier to hire the skilled professionals needed to implement and maintain commercial databases -- particularly the most commonly used ones -- versus open source options.

"If you have a large-scale organization that needs support and constant product updates that are secure and in compliance, proprietary tools are great for all that," said Eric Chi, a director in the technology practice at consulting firm West Monroe.

Those are important considerations for organizations looking to quickly advance their use of the data they generate and collect.

You're not having to test and throw things in and out [of proprietary systems], you're not managing a development team and code updates, you're not forced to build a software business inside your otherwise traditional company.

With proprietary systems, "the kinks have been worked out," Srivastava said. "You're not having to test and throw things in and out, you're not managing a development team and code updates, you're not forced to build a software business inside your otherwise traditional company. You have someone else doing all that -- doing all the worrying and working to get everything in the system right."

However, proprietary database management software can come with some potential downsides, he and other consultants said.

To start with, enterprise teams can't innovate on proprietary code and instead must rely on the vendors to keep pace with the innovations necessary to succeed in a rapidly evolving digital landscape. Proprietary software also costs more in upfront fees than open source options. Additionally, SMBs without complex IT environments, large data stores and big data projects might find proprietary systems have a higher total cost of ownership than open source ones.

There's also the chance of getting locked in with one vendor, with the cost and challenges of switching to another proprietary database overwhelming the benefits of making the switch.

Open source database management software pros and cons

In contrast to proprietary DBMSes, open source database technologies are released under licenses that let users deploy the source code to develop their own systems and also update, change and modify it for their own needs. That flexibility enables the creation of data management systems that fit each organization's unique needs, McKnight said.

"With open source, if you're so inclined, you can create forks of your own in the code. For some, that might matter," he added.

Open source software has the potential to be less expensive to use, particularly when starting out, as there are no licensing fees. That means organizations often can test open source options more easily than proprietary ones, enabling them to run a proof of concept or pilot project before deploying an open source database more broadly -- or upgrading to paid enterprise versions offered by vendors, sometimes with added proprietary features.

Additionally, open source gives database administrators and developers the ability to innovate on the source code and to draw on improvements to it that others in the open source community make. "With the open source community, you have more people contributing to lines of code, so you're going to get more innovation," Srivastava said.

Those benefits have many enterprise IT and data management leaders looking at open source, Yuhanna added.

"Over the past few years, Forrester has been receiving a lot of customer inquiries about open source, and more organizations -- even Fortune 100 companies -- are looking at open source," he said. "We see open source playing a big role in data strategies. What we've seen is that open source tools can lower costs, help avoid vendor lock-in and future-proof your architecture."

On the other hand, open source databases -- like their proprietary counterparts -- come with possible drawbacks.

Open source software is generally harder to integrate with other data management tools than proprietary alternatives, Yuhanna said.

Furthermore, organizations looking to take full advantage of open source databases on their own need technologists with the skills required to deploy systems and maintain, modify and improve the source code. Those staffers must keep up with changes to the source code by others, and they might be required by the software license to contribute any modifications to the open source community. They also have to be capable of doing all that work without the 24/7 customer support that typically comes with commercial software products unless their organization pays for support from a database vendor.

With open source, you're generally getting the latest and greatest features out there, but the downside of that is those features are untested -- the community is testing them as they use them.

"With open source, you're generally getting the latest and greatest features out there, but the downside of that is those features are untested -- the community is testing them as they use them," Chi said. "And when you have questions or need support, there are questions about who you rely on."

Chi also noted that while an open source database can be the lower-cost choice when starting out, its total cost often exceeds proprietary options when calculated over the long term or when an organization scales up its database operations. Additionally, data management teams often find that shifting from open source software to a commercial product as their needs grow can "be a very costly process," he said.

All these considerations can make an open source database more work and result in a higher cost than the value an organization would get out of it.

As McKnight explained, open source might indeed let organizations create their own forks in the source code, but that opportunity might come with more challenges than such customization is worth, as it "nearly makes the enterprise into a mini software company," he said.

The future is a mix of database technologies

IT and data management leaders in an organization don't need to make a one-or-the-other choice between proprietary and open source databases for all use cases. Rather, they can use proprietary DBMS software for some data needs and open source technology for others.

Chi, for example, said an organization might be large and complex enough to need a proprietary DBMS for most of its needs but opt for open source for a niche use case or one that requires highly customizable software.

Similarly, he said a small company without any complex data requirements that had been a good candidate for an open source system for cost reasons might find a proprietary DBMS from a vendor offering a per-user or per-usage fee structure to be a more cost-effective option.

Organizations might also go with commercial database providers for data management and storage but use open source options for other purposes, such as Apache Kafka or Apache Spark for near-real-time data processing, Apache NiFi for data orchestration and Apache Airflow for workflow management.

Some organizations also might want to use open source databases for pilot and proof-of-concept projects and then move to commercial choices when scaling up for full deployments.

Given the endurance of both open source and proprietary databases as viable options, and their possible benefits and drawbacks, McKnight said data management leaders must be prepared to evaluate the options and choose which is better for their organization for each use case at any given point in time.

"The decision should be a conscious one," he said, "so you need a goal to move toward. That goal may change over time, but you should still have one to help guide your decisions on how to platform new initiatives."

Mary K. Pratt is an award-winning freelance journalist with a focus on covering enterprise IT and cybersecurity management.