Moving to APIs may require modernization efforts across legacy platforms such as those supporting COBOL applications. This may be a daunting task, but there are workarounds in which the legacy applications do not have to be replaced, Horn de Oliveira adds.
Ultimately, the key will be “committing to a sane subset of technologies that have broad enough application so enterprises can use them in a variety of situations, while being effective enough at tackling them specifically,” says Eliot Horowitz, the CTO and co-founder of MongoDB. “That’s actually why relational databases have survived this long—they have broad application, as long as you keep gluing more and more translation on top of them. I think it’s easy to go from one extreme to the other—one database for everything to a niche database for everything—but I don’t think that will be the best strategy. I think there will be a handful of technologies that an enterprise will lean on, covering all of their needs.”
Ultimately, is it a bad thing to have so many different data environments? Not necessarily, Imhoff believes. “It’s never bad to be disruptive,” she says. “It’s just that a lot of people don’t like change. And if that’s the attitude you go in with, then this is a very bad period. On the other hand, if you look at it as being disruptive, which means there’s an opportunity, then it’s exciting, it’s interesting, and it’s good.”
The key is to keep an open mind and open architecture to emerging solutions that provide the best capabilities. “At PayPal, we have petabytes’ worth of information across multiple data centers,” says Bayhan. “Our chief concerns, given that we have 137 million active customers in 193 markets, are security, availability, and scalability. We scrutinize every technology that we consider for our infrastructure.”
The focus should be less on diversification of the database platforms, and more on the use case, Attal agrees. “The use of multiple database types is not a serious issue for enterprises as long as there is a valid business case for use and IT has the capabilities required to deploy, maintain, and govern the new database. In other words, does the new database technology provide sufficient business and/or operational value to warrant investment in its use? The general wisdom of ‘use the right tool for the job’ applies.”
That doesn’t mean letting go of traditional technology entirely, Imhoff continues. “Can you somehow bridge this ever-changing world with your traditional world?” she asks. “Yes, there is a bridge. I think we’re finally beginning to figure out that there is still a need for the traditional enterprise data warehouse and its various dependent parts. But there’s also this new world of investigative computing platforms. Once we figure out what’s truly of value to the organization, that’s where we start to pull it all together and begin to apply architectural principles.”
Image courtesy of Shutterstock.