Apache CouchDB

It's freely available to download in several versions including for Linux, Windows and Mac. It uses an HTTP interface and is for all intents and purposes relatively easy to use with the built in interface called Futon. So it's much quicker to browse or create new content or data compared to using a third party client to search SQL effectively. The map reduce feature is pretty cool in how it allows you to process massive datasets in a cluster or a grid and store it either as a structured database or unstructured file system. It's great you can store and replicate data across nodes is good. CouchDB is a challenge if you enjoy that and have never installed it before. Personally, I like it when you reach the point when it says 'Apache CouchDB has started, time to relax.' because until that time it's a serious unknown with all of the various elements which it takes to setup and extract initially so this is satisfying attempting to put my most, sincere positive spin on it. As it's a non-relational based database you have cool relevant capabilities like merging or the ability access multiple databases like on a smartphone and another somewhere else. The multi-version concurrency feature is good too how it helps you resolve any conflicts as part of its core functionality. I also like the multi-master replication allowing you to scale across nodes to increase performance. The documentation is decent and there's an issue tracker you can submit to as well as a wiki so that's all good. And if you Google 'CouchDB tutorial' you will find a lot of third party resources and guidance if you want to take steps yourself. Or you can just hire an expert though this may nullify the cost savings of using open source. In this case, you may consider using something like ElasticSearch if that's all your requirements are and it's relatively simple to use.

Perhaps it's not as much of a search tool as it is best at querying which is not quite the same thing. You must have well defined queries of your attributes in order to utilize CouchDB to it's fullest potential so it's not like Googling your cluster which would be cool. The filters you can run simultaneously are limited in CouchDB for example others solutions may allow you to utilize metadata, various ranges, full text and spatial requirements all at once and it doesn't appear to be possible from CouchDB.

This is a solid alternative to SQL with it's schema-less JSON data stores. Your ability to organize your massive amounts of data in a centralized place while denormalizing for a glance at your attributes all in one place is great and simpler than some other ways of going about it. Though there are perhaps easier ways to have search functionality rather than querying such as Hive I believe, or ElasticSearch. For the price you can't complain about CouchDB. Free for all forever sounds good to me, though perhaps I would suggest utilizing it in addition to a paid solution to have the option for resolving any sort of time based issue perhaps quicker such as ElasticSearch. If you're slicing your data really specifically and have well defined attributes you can utilize CouchDB just as effectively. I believe CouchDB is an excellent secondary data store to have as an option certainly.