Competition generally drives innovation -- that's how capitalism works. But up until now, despite its overall worthiness, MongoDB had no credible competition from other document databases. That changed on Wednesday with the general availability of Couchbase 2.0, which adds document database capability to the leading key-value pair database.
Couchbase was sort of a fork of CouchDB, the Apache project founded by Damien Katz. He discovered what took me much longer to discover with my own project: Apache is a tough place to develop software and nearly an impossible place to develop commercially viable software.
[ Also on InfoWorld: The time for NoSQL standards is now | Keep up with the latest developer news with InfoWorld's Developer World newsletter. ]
Created as a more scalable, production-ready solution than CouchDB, Couchbase was both a step forward and a step backward. Key-value stores are great for certain types of problems, especially those that require low latency. As it turned out, the market for key-value databases is much smaller than the market for the almighty document database in which Katz & Co. had their roots.
Couchbase has landed great clients -- LinkedIn, Orbitz, and Cisco, to name a few -- which have provided some important case studies. At a NoSQL conference, you expect to see typical early adopters like social gaming companies, but blue-chip tech companies made their presence known at the CouchConf in San Francisco. For example, there's Orbitz, which has the kind of traffic and occasional data challenges that most people only talk about. Stephen Young of Orbitz gave an impressive talk with compelling statistics about the performance of Couchbase and Orbitz's overall experience.
Couchbase, as a document database, should expect to see even more uptake. This will happen because Couchbase intelligently handles so many issues common to distributed databases.
Compared to MongoDB, Couchbase handles concurrency better, and while actual high-end usage will tell, it architecturally looks like a more scalable model. Couchbase takes the concurrency of memcached and combines it with a storage engine inspired by Apache CouchDB, but rewritten in C. MapReduce functions in JavaScript are testable in a nice GUI admin panel. The cluster manager handles automatic rebalancing, sharding, and cross-data center replication, so scalability is a snap. Data sharding uses hash values, solving the issue Foursquare had back in 2010.
On the other hand, Couchbase is written in C and Erlang whereas MongoDB is written in C++. Inevitably, you'll need more memory for Couchbase as compared to MongoDB. Couchbase has some interesting advantages when approaching large corporate environments. It is also helpful to have that key-value store supported from the same product. You can pitch as either a database and sit next to Oracle, or you can pitch as a middleware technology akin to a distributed cache like Terracotta and sit it next to the app server. This flexibility can give Couchbase a "backdoor" opportunity.
The new Couchbase should have an easier time getting into the enterprise, but it has a lot of catching up to do to get into the cloud and to capture the hearts and minds of developers. Mongo is very developer friendly and well-ensconced in that community.
On the other hand, with features like out-of-the-box datacenter replication controlled from a pretty Web menu, Couchbase offers much to love. Web-based administrative menus and such seem like trite features to harp on until you're actually selling this stuff to management: Show them a command-line interface and talk about how you have to write another project to get the data out after you adopt this thing -- congratulations, you're facing an uphill battle. Show them the Couchbase GUI interface, which at least demonstrates that the vendor has put some thought into BI, reporting, and so on, and you have a better story.
It's now obvious that 2013 will be a year of NoSQL war. Players will escalate their jockeying for supremacy. Gone will be the NoSQL peace, love, and understanding, where we're all in this together -- only Larry is the enemy. Instead, we'll have a popcorn-worthy conflict to watch.
This competition is no longer strictly about mind share among developers and early adopters. This is about the hearts and minds of enterprise IT management. The result will be better administration tools, more partnerships, and a growth spurt to the maturity necessary for more conservative enterprise customers.
Bring on the document database war! I want more!
This article, "Couchbase 2.0: This means war," was originally published at InfoWorld.com. Read more of Andrew C. Oliver's Strategic Developer blog, and keep up on the latest developments in application development at InfoWorld.com For the latest business technology news, follow InfoWorld.com on Twitter.