It’s the end of NoSQL as we know it (and I feel fine)

Last week I tweeted that this week was shaping up to be a watershed week in the history of NoSQL. I was referring, of course, to MongoDB launching 3.0 and DataStax acquiring Aurelius – although more specifically what the context of these two announcements tells us about the future of NoSQL.

While each of these announcements could be considered significant in its own right in combination they suggest a new stage in the evolution of NoSQL and a clear signal that the future of NoSQL will be driven by database products that support multiple data models.

When we formally started covering NoSQL in 2010 it made sense to divide the various projects into four groups: key value stores, distributed (wide) column stores (or BigTable clones), graph databases, and document-oriented databases.

By early 2013 it had become obvious that there was another emerging category: multi-model databases.

Multi-model NoSQL databases have therefore been around for several years but while we have seen growing interest in these multi-model databases, in terms of widespread adoption they still lagged behind the early specialist NoSQL databases. That’s what makes the recent announcements by MongoDB and DataStax so significant.

    1. Along with releasing version 3.0 of its document database, MongoDB also began to share (at least with us) its long-term multi-model vision for MongoDB, explaining how the pluggable storage engine architecture could enable the database to support multiple data models – such as key value, graph and relational.
    1. Meanwhile DataStax described how its acquisition of Aurelius will see it developing a graph database to complement Apache Cassandra’s wide column key value model, and explained its multi-model strategy.
  • Multi-model momentum may have been growing for years but the fact that the commercial providers behind the two most popular NoSQL databases have detailed their plans to go multi-model confirms that the multi-model approach is the future of NoSQL.

    Indeed, since we expect to see similar moves from other NoSQL players it will become increasingly difficult to divide the NoSQL space in terms of key value stores, wide column stores, graph databases, and document-oriented databases. Instead it makes sense to divide the NoSQL projects in terms of whether they are single-model or multi-model.

    451 Research clients can read more about our perspectives on MongoDB’s strategic direction, as well as DataStax’s acquisition of Aurelius, and the wider implications for the NoSQL sector.

    Tags: , , , , , , , , ,

    1 comment so far ↓

    #1 Luca Garulli on 02.04.15 at 8:45 am

    I can’t agree with you more: Multi-Model is the NoSQL 2.0. OrientDB was the first product to have this vision 5 years ago when we built a true Multi-Model engine that supports Graph, Document, Key/Value and Object models.

    Now my question is: how many years will it take for these “new” Multi-Model NoSQL competitors to have a rock-solid multi-model engine?