Click to learn more about author Kevin Kline.
Tech pros must balance productivity with cost and complexity every day, and those who work with database platforms are further challenged as open-source databases become more commonplace.
Additional complications range from the transition to remote work to the rising demand for data processing, compliance with GDPR, and ever-expanding data volumes. Add to this the hybrid IT reality and throw in yet another variable: the growing number of databases tech pros must manage.
Nearly one-third of tech pros surveyed in the latest Query Report say they’re responsible for more than 300 databases at their organizations. And most respondents say half or more of their organization’s databases are considered critical to their business.
Further still, companies concurrently work with a wide variety of database platforms. The rise in this widening variety and commensurate lessening of brand loyalties is driven by the need for enterprise IT to deploy applications at an ever-quickening pace. Consequently, many enterprise IT teams are choosing to buy applications rather than to build their own. These third-party applications frequently deploy with a preferred data platform, which can be relational or non-relational (the so-called NoSQL data platforms), and they can also be either entirely open-source or commercial.
Commercial platforms still dominate – especially for business-critical applications – but open-source databases are on the rise. In previous eras, few enterprise IT teams were willing to inject risk into their internal systems by using an open-source data platform or a database management system from a new startup. Those days are over. We’re at the beginning of a new era of mix-and-match data platforms, but there are critical provisos to keep in mind.
Don’t adopt an open-source platform before investigating all your options first.
You Can’t Beat Free – or Can You?
The advantages of open-source databases are many, with the salient one being obvious: You can’t beat free or low-cost. Without the (often expensive) licensing fees attached to commercial software, IT budgets are unencumbered by contractual payments. With the savings, IT departments have enormous flexibility, can experiment, and can move as quickly as market forces demand.
Because the open-source community doesn’t grant licenses based on fees paid, small companies have the same access to innovative features as multinational ones. Startups and other small businesses arguably have an advantage over larger companies. They generally make decisions faster because there are fewer layers of management to reinterpret and reprioritize the factors needed for quick decision-making.
In the past, publicly traded companies were reluctant to deploy open-source database platforms, despite the savings on the licensing fees. They were concerned with legal ramifications and accountability: When something went wrong, who was going to fix the problem? Would they have to wait on a patch from a volunteer on the other side of the world?
Whether or not you should adopt open source comes down to cost versus value. An enterprise may opt for a commercial tool capable of saving them thousands of dollars a year because of efficiencies and ready support. The licensing fee – even if it’s six figures – may not seem so high because, well … it works. And when it doesn’t, you have direct contractual recourse to fixes.
Others may opt for a low-cost database platform, willing to work with a platform not as sophisticated as a commercial one to benefit from the cost savings. Sometimes, the IT team may simply choose to use only the most common and well-vetted features of an open-source database platform to avoid undue risk. Additionally, it’s possible to contract with commercial enterprises such as Red Hat, Debian, Percona, and others who manage many of the problems and bugs potentially arising in an open-source database platform.
Adopting open source also means you won’t be alone: According to the aforementioned report, 43% of tech pros say they’re currently running MySQL or MariaDB. Additionally, 18% say they plan to adopt MySQL, MariaDB, or another open-source database platform in the next three years.
But before running into fresh territory, remember this added drawback: The tooling usually isn’t sophisticated, and the platform likely doesn’t have a long history in the marketplace to support additional tooling you might otherwise take for granted with commercial database platforms. Open-source platforms often call for knowledge on how to tinker with them to get what you want if it’s anything beyond the basics. Although many people know the open-source platforms (recent college grads have often studied them in school), this downside is a concern for the small and medium businesses in need of strong support and tooling.
Meanwhile, enterprises are looking to open-source databases for basic database applications – or at least weighing the pros and cons. They have the on-hand expertise with database management. Ultimately, many enterprises decide to wait for better tooling before committing to a frontline adoption of an open-source database platform.
First, Investigate Your Options
Before jumping into open-source databases, database pros should investigate why and when these platforms make the most sense for their organizations. Many times, the initial steps in an investigation about whether to use open-source databases focus only on the performance and workload capabilities of the platform. But this isn’t enough.
You should also ensure you’ve covered all your bases outside of performance essentials, especially regarding security and compliance (and, on occasion, cost and licensing). Remember, the new laws on compliance have teeth, and security should never be an afterthought.
Admittedly, finding the time to investigate these matters can seem like an insurmountable problem. After all, our report found database professionals work with diversified platforms of more than 300 databases. Just keeping this many databases up and running, resolving problems, and fighting fires is more than a full-time job.
In fact, our report also found maintenance work accounts for a substantial portion of the database professional’s day: One-third of tech pros surveyed stated much of their days were spent in maintenance. This leads to less time for investigating open-source database platforms or, for that matter, any sort of analysis potentially leading to productivity gains and cost savings.
A Monitoring Mindset
Tech pros can save time spent on maintenance when they implement appropriate automation and monitoring tools. Database monitoring tools, in this case, can foresee problems before they reach criticality and can even respond to various alarms in real time. Automation, on the other hand, frees up time to focus on proactive database performance management. While database professionals upskill, innovate, and investigate their open-source options, monitoring tools can run in the background, further leveraging their ability to manage multitudes of databases.
One thing I’ve seen over many decades of experience is enterprises with a monitoring mindset are much more successful than those without one. When top shops get a new database platform, they don’t even consider moving it into production without a monitoring method ensuring it can’t break without them knowing about it. They make sure it doesn’t run amok, taking all the processing cycles on a particular server. Doing this helps enormously as tech pros continue to add more things to their plates.
Also, when they adopt practices and processes – actual workflows – no one makes knee-jerk decisions based on office politics. I’ve seen it countless times – the database professionals have the power, so the development teams have to follow their lead. At other shops, the dev teams have more political power, and the DBAs must keep up. The result is many fiefdoms and no real king, a recipe for chaos.
Open-source databases are here to stay. Before deploying one, however, investigate your options thoroughly (remembering the consequences for a wrong choice are real). Then, when deploying the new database platform, be sure to deploy automation and monitoring along with it. And throughout, have rules of deployment based on clear business goals – take the politics out of the mix.
Sooner or later, you’ll likely have to work with an open-source database platform. The landscape is improving, and once the groundwork has been laid, proper tooling will follow. If you haven’t already, start to learn about open-source database platforms and get proactive.