NoSQL databases are “a complete game changer,” according to database expert Michael Franklin. But that hasn’t stopped traditional enterprises from persisting with traditional databases. Part of this stems from comparatively immature tooling for NoSQL databases. Part of it also derives from enterprise workloads remaining largely transactional in nature.

In a sign that the market is seeing a significant shift to NoSQL, however, the Forrester Wave for Big Data NoSQL, Q3 2016 declared that “NoSQL is not an option–it has become a necessity to support next-generation applications.” The question increasingly isn’t whether “enterprises of all types and sizes are embracing NoSQL”–they are–but rather whether they’ll run those databases in the public cloud or on their own servers.

You need NoSQL

There are plenty of reasons to adopt NoSQL databases, but Forrester calls out the most critical:

NoSQL is not an option–it has become a necessity to support next-generation applications. And increasingly, enterprises of all types and sizes are embracing NoSQL to support their business technology (BT) agenda. A key strength for NoSQL is the ability to support scale-out architecture leveraging low-cost compute servers that are clustered to deliver performance of large, high-end SMP servers. In addition, its flexible schemaless model offers the ability to store, process and access any type of customer and business data.

SEE: The best of big data NoSQL: MongoDB, Amazon, and DataStax top Forrester list

Small wonder, then, that 61% of enterprises surveyed by Forrester have already implemented NoSQL databases or have near-term plans to do so:

Also not surprising, the most popular NoSQL databases, including MongoDB and Cassandra, top Forrester’s Wave:

These databases provide the broadest utility without sacrificing performance. Lurking just behind MongoDB, however, is AWS’ own NoSQL database service, DynamoDB. Microsoft’s Azure DocumentDB has also been growing in popularity, though it still has plenty of catching up to do.

Get your head in the clouds

Despite the success of Amazon Web Services and Microsoft Azure, we’re still a long way from most enterprise workloads running the cloud. That said, there are clear indications that we’re rapidly moving in that direction.

For one thing, the growth in public cloud VMs (20X) trounces that of private cloud VMs (3X), according to Gartner data. The primary reason for this outpaced growth in public cloud is business agility.

As AWS product chief Matt Wood told me, “If you buy infrastructure it’s almost immediately irrelevant to your business because it’s frozen in time. It’s solving a problem you may not have or care about any more.” Smart enterprises optimize for agility, which increasingly means the public cloud, and not a glorified private data center masquerading as “private cloud.”

In other words, NoSQL delivers one side of the business agility equation, allowing for disparate data types at high velocity and volume. Public cloud takes care of the infrastructure side of the equation, enabling enterprises to grow or shrink resources according to data demands.

SEE: NoSQL keeps rising, but relational databases still dominate big data

Gartner’s newest projections suggest that a decent chunk of that NoSQL market will be filled by public cloud providers like AWS and Microsoft Azure, not to mention MongoDB with its new Atlas service, as it and others try to satisfy a growing desire to run databases in the cloud. In fact, according to Gartner, by 2020, public cloud DBMS licensing will account for 9-10% of the DBMS market.

In short, as the world embraces NoSQL for next-generation applications, and turns to the public cloud to run those applications, we’ll increasingly see NoSQL database cloud services as the platform of choice for tomorrow’s data.