Five Ways Open Source Databases Are Best for Business
Today 78% of organizations run part or all of their operations on open source software, a figure that has nearly doubled since 2010. And according to ranking site DB-Engines, six of the top 10 databases are open source, and the top eight non-relational technologies are all open source.
So why do so many organizations standardize on open source? Why do 66% of organizations look to open source before considering proprietary software alternatives? When it comes to databases, it turns out that the most important criteria are likely to be better addressed by an open source product.
Which criteria are we talking about? Factors such as the quality of the technology and its feature set; the availability of support and technical experts who know how to use the product; whether it integrates with the rest of your technology standards; to what extent the technology will evolve to meet future needs; and last but not least, cost.
The license model on its own makes no guarantees, but the database industry has learned over the past 15 years that the best way to address the criteria that matters most is to release the technology as open source.
Here are five reasons why virtually every important new data management technology over the past 15 years–from Hadoop to MongoDB to Spark– is open source.
It is true that many people like open source because it’s free, but being free doesn’t make it better. The thing that makes open source databases better for modern work loads is quality.
If you write code and put it out to the world you are accountable for it. People can inspect and admire it, but they can also be ruthlessly critical. Hopefully they make their own contributions that improve the quality of the project. As a result, the overall quality of open source projects is consistently higher than that of proprietary products, according to the recent Scan Open Source report from code tester Coverity.
To quote the researchers: “Open source code defect density improved from 0.66 in 2013 to 0.61 in 2014, while commercial code defect density improved from 0.77 to 0.76.” The study also found that open source project Linux is, year in and year out, the highwater mark for code quality and defect density.
When large companies deploy databases at massive scale they identify issues that would never appear during QA – with open source those organisations can examine the code and then contribute patches back into the product long before more typical enterprises are impacted. The web Titans – like Google and Facebook – build their businesses on open source, and now contribute their own innovations back as new open source projects for others to use.
The open source development model has become so successful that Blackduck’s research revealed 39% of enterprises expect to start an external open source software project in the next two to three years, compared to just 9% last year.
Community inspection and contributions are perhaps most important when it comes to security. Developers evaluate the code base to ensure security best practices are followed and that no backdoors exist. The community keeps you honest and ultimately drives the quality of the product to a level that no single organization could achieve on its own.
In the past, open source has been accused of simply replicating the features of proprietary products. That is something that may well have been true 10 to 15 years ago, but certainly isn’t true today.
Consider some of the most recent innovations in database technology – the use of JSON data structures and dynamic schema, the ability to shard the database on low cost commodity nodes, and the ability to deploy across global data centers. These features are now viewed as critical for modern workloads, and they first appeared in open source technologies such as MongoDB. The power and speed of community innovation is simply much faster than any single entity.
Now the tables have turned, so to speak, and the proprietary relational database vendors are trying to shoehorn open source innovations into their legacy architectures. In multiple surveys where users cite a broader feature set, ease of deployment, scalability, and their newfound business agility as the key advantages of open source software over proprietary alternatives.
3. Access to Talent
The biggest challenge I hear from IT leaders is access to top technical talent. Finding great people is really hard. That’s why widespread adoption matters. You want a technology that people know how to use and want to use. In fact, over 50% of companies surveyed by Blackduck state that using open source software helps recruit new talent to the business.
If your application is in production and your database administrator suddenly leaves, you want to be confident that there are other people in the world who can take on the work. That strength of community and engagement that comes with an open source project provides a broad talent base. At MongoDB over 300,000 people have registered for our public training, and over 10,000 people download the software every day.
4. Integration with Your Standards
The momentum around popular open source technologies spreads to create a broad ecosystem of complementary technologies and standards. Because open source software is widely used and easy to acquire, other technologies tend to develop their own integrations. As an example, MongoDB Inc. provides drivers for 11 programming languages while the wider MongoDB community provides more. There are also many community developed connectors for search engines, message buses and data visualization and analytics tooling.
No software exists in isolation. You want your technology choices to work seamlessly with the rest of your stack. That’s more likely to happen if the database is open source.
5. Long-Term Viability and Avoidance of Lock-In
All organizations want to build on software that will remain stable and continuously improve. With proprietary software you’re at the whim of the maintaining entity. As we’ve seen many times in the world of software, companies come and go–sometimes through acquisitions, and sometimes as a result of failed business models. In these cases your options with proprietary software can become very limited.
What’s more, vendor behavior can sometimes be….let’s just say “less than seemly.” The problem with proprietary products is that you have few options. The least appealing but most likely option is wholesale migration, which is both costly and risky, especially if you have to achieve it before your license expires. In the case of open source software, if you are not happy with the direction the vendor is taking, simply fork the code. Lock-in eliminated in one Github commit.
If MongoDB Inc went out of business tomorrow, the project would continue on through the efforts and stewardship of the MongoDB community.
About the author: Kelly Stirman is the VP of Strategy at MongoDB. For over 15 years he has worked at the forefront of database technologies. Prior to MongoDB, Kelly served in executive and leadership roles at Hadapt, MarkLogic, Oracle, GE, and PricewaterhouseCoopers.