So check it out, man, there’s this document database, right? It’s called Couchbase, and they’re like adding this new columnar side-car thing to boost analytics performance for users who want to dive deep into their real-time data.
They announced it at AWS re:Invent 2023 in Las Vegas, and this new service brings in a columnar store and data integration into the Capella Database-as-a-Service (DBaaS) for stuff like customer profiling, and get this, special offers.
Now here’s the kicker – this in-memory analytics system only comes as a package deal with the main DBaaS, but it’s all good because it supports tools like Tableau and PowerBI for analytic development and visualization. Oh, and they also rolled out a new conversational coding tool called Capella iQ, designed to help developers use natural language interactions with ChatGPT for coding in SQL++. So, you know, lots of cool stuff on the horizon.
And this new service’s gonna be in private preview next year and is expected to go public in the fall.
Now get this – I stumbled across this other database called MongoDB, and they added analytics features last year. They created this column store indexing to help developers speed up a bunch of common analytical queries without having to mess with the document structure or move data to another system. And when I talked to some analysts, they said MongoDB might be better suited for simple queries, not so much for complex modeling.
But, here’s where it gets interesting. Couchbase is saying they’re doing things differently, you know? They’re making sure to point out that MongoDB created a duplicative indexing structure, basically doubling up on the data that’s already living in its singular storage engine, WiredTiger. Couchbase, on the other hand, claims their approach allows both the columnar and document engines to work and scale independently in the same cluster. And of course, I reached out to MongoDB for the scoop on all this, so we’ll see what they say.
Then there’s this guy, Chris Bridgland, who’s the senior director of solutions engineering and customer success Europe. He’s saying that the columnar store in Couchbase supports Avro files, which are super popular in the telco industry. And he’s pointing out that Couchbase creates the index and the schema when it’s reading the data, instead of having to set it all up beforehand. And, let me tell you, they’re seeing some pretty solid improvements in performance during testing.
And finally, this industry analyst, Doug Henschen, he thinks the new columnar database move is gonna be super appealing to Couchbase customers who want more analytical capabilities alongside the platform’s transactional features. But the thing is, he doesn’t really see it as a head-to-head competition with MongoDB. He’s saying it’s more about these database providers helping customers build next-gen apps with analytics. It’s gonna be tough to really compare the performance until customers start building out these proof-of-concept systems, you know what I mean?
So that’s the deal, folks. Looks like this whole transanalytical capabilities thing is turning into one wild ride, and we’re just strapped in for the journey.