Bring Sanity to Managing Database Proliferation
There seems to be no end to database proliferation — and for good reason. Databases are critical to any modern organization for providing essential business insights and building apps that power the digital economy.
Developers embrace different databases for their many different attributes: SQL; NoSQL; open source or high-volume legacy databases; databases optimized for speed or scale; transactional and relational databases. As a result, according to Gartner, the market for database management systems (DBMS) has shot up to nearly $80 billion, from $38.6 billion in 2017.
While there are only a few large players in the legacy database space, IT leaders are faced with managing the ever-growing number of databases that are essential building blocks for modern applications. It’s just the reality of our cloud- and DevOps-oriented world.
There are currently 399 different database systems. As developers use their preferred method for charting data, working with increasingly large data sets and more, the holy grail is finding ways to consolidate how you run them all. Why? Each one installs in different ways, manages data in different ways and ultimately scales in different ways. It can be a nightmare if you let it be. But you can stay out of the trap of trying to manage all these individually.
Getting a Grip
How can you avoid being a victim of the bow wave of database proliferation? Recognize that you can allocate your resources in a way that benefits both your bottom line and your stress level by consolidating how you run and manage modern databases.
Investing heavily in self-managing the legacy databases used in high volume by many of your people makes a lot of sense. Database workloads that are typically used for mission-critical transaction processing, such as IBM DB2 in financial services, are subject to performance tuning, regular patching and upgrading by specialized database administrators in a kind of siloed sanctum sanctorum.
Many organizations will hire an in-house Oracle or SAP Hana expert and create a team, for example, or even seek someone with special expertise in Amazon RDS, a heavily used cloud database.
But what about the 40 other highly functional, highly desirable cloud databases in your enterprise that aren’t used as often? Do you need another 20 people to manage them? Open source databases like MySQL, MongoDB, Cassandra, PostgreSQL and many others have gained wide adoption, and many of their use cases are considered mission-critical. But these modern-day tools use cloud technologies and modern development techniques.
These modern databases follow a distributed model versus the legacy database monolithic, siloed model. In the financial services world, for example, it’s the difference between core banking services and mobile banking.
Legacy banking services have dedicated infrastructure for storage, compute and networking. Mobile apps rely heavily on microservices, containers and Kubernetes.
Born-in-the-cloud apps and databases are rapidly changing. They have the advantage of being agile and the ability to add new capabilities quickly. But teams that manage them have a lot on their plates, managing not only multiple database instances but a wide variety of database types too.
Solutions like the database-as-a-service platform (DBaaS) that Portworx provides offer a way to make these services scalable, highly available and secure in a self-service, on-demand way. With an effective DBaaS platform, small teams can support a large number of users, providing reliable operations and an easy small-service user experience.
Scalability is essential in our fast-growth era. The ease with which DBaas users can add nodes, memory and storage resources can substantially reduce the administrative burden on IT departments and enable a more frictionless experience for DevOps teams to perform tasks more quickly and efficiently.
Getting a handle on this now is important as this need becomes more pressing. As we head into 2023, a DBaaS platform should find a place at the top of IT leaders’ wish lists.
