It is increasingly clear that SQL databases are not the perfect answer to every data storage and retrieval problem. The industry has reached into the past and brought back pre-SQL databases back to the forefront to address need to address data that really isn't happy being forced into a "first normal form" for processing. Membase, a supplier of a No-SQL, memory cache database, and CouchOne, the supplier of a document database, have merged to create Couchbase with the plan of addressing the rapidly increasing need for management of non-structured or non-SQL structured data.
What Couchbase has to say about this announcementCouchbase is the name of both our new company and our new product family. Apache CouchDB document database technology is at the core of our combined solution. Membase, with its integrated memcached caching technology, adds technology enabling dynamic cluster elasticity and sustained low-latency, high-throughput data operations. Couchbase becomes the only document database capable of safely storing your data whether stored on a single server, or spread across hundreds. Couchbase: Setting out to redefine the NoSQL landscape.
Only Couchbase technologies scale from AOL-sized data center clusters all the way to smartphones. Only our solutions meet the needs of developers and conform to the mission-critical demands of ops teams. And as users increasingly access applications via smartphones and tablets, we alone are able to support the entire range of interactivity and connectivity.
Snapshot analysisI've spoken with Membase in the past (see Membase practical approach to No-SQL for more information.) CouchOne, on the other hand, was new to me. It appeared that they were doing their best to deal with the growing amount of data organizations are managing and the fact that a SQL database, often prescribed by database companies, is not always the right tool for the job. We enjoyed a long conversation about database software, what it does and why there have been so many different types of database products over the years During the discussion, I was reminded of my years as a software engineer working on an Automated Library Information System (ALIS) which was built using Meditech's Interpretive Information System (MIIS), a form of MUMPS.
MUMPS was fantastic at managing structured, non-structured or semi-structured data in an environment requiring fast retrieval and storage. It was not all that good of a tool for environments that needed to do batch processing of every record due to the structure of the database. Used properly, it could best just about any other database product at the time.
I was also reminded of a different system, PICK, that did something similar, but used different technology.
In the end, we all agreed that the move to memory cache-based data management, sometimes called shared cache, memory virtualization and a few other names, can be a great solution to problems that just aren't addressed well by a SQL database.