NoSQL databases do away with fixed schema, which can be beneficial when developing applications with changing requirements. What separates these databases from traditional databases has less to do with SQL and more to do with departure from relational models.

There are several types of non-relational databases that fall under the NoSQL. These include key-value stores, document-oriented databases, columnar databases and graph databases - each with its own data models, scaling strategies and use cases.

In a key-value store, individual records amount to an arbitrary lump of information, indexed by a key. These systems typically do not interpret the data, leaving that function to the application. Riak(supported by Basho Technologies) and Oracle's Berkeley DB are popular key-value stores.

In a document store, record comprise documents that consist of a variable number of named attributes of various types, such as integers, strings and nested object. Document-oriented databases tend to recognize the structure of the data they store and have more querying functionality than key-value stores. 10gen's MongoDB and Apache CouchDB(supported by Couchbase) are example of popular document stores.

Extensible-record stores, also known as wide-column stores, provide a data model similar to relational databases, but with a focus on organizing data into column families(rather than tables). Apache Cassandra(supported by DataStax) and Apache HBase(supported by Cloudera) examples of popular extensible-record stores.

There are a number of SQL-like querying languages that have sprouted up to offer higher-level data access. These include Google's GQL for its AppEngine PaaS(platform as service). MongoDB's Mongo Query Language, Cassandra Query Language and the nascent UnQL(Unstructed Query Language). For Apache Hadoop-based systems, Apache Pig and Appache Hive offer two separate route for working with data from a higher level.

To understand the broad categories of NoSQL data stores, this paper by Rick Cattel, the former Sun Microsystems database architect, is very helpful: http://tinyurl.com/3gh98mw

As a researcher at the University of California, Berkeley, in the early 1970s, Michael Stonebraker co-created the Ingres and Postgres technology that underlies many leading relational databases today: Microsoft Corp.'s SQL Server, Sybase Inc.'s Adaptive Server Enterprise, Ingres Corp.'s eponymous product, IBM's Informix, and others.

But Stonebraker now argues that relational databases, also known as RDBMSes, are "long in the tooth" and "should be considered legacy technology."

In an entry Tuesday at a new blog, The Database Column, Stonebraker also argued that today's relational databases lag badly in performance behind a new wave of databases that flip database tables 90 degrees.

Column-oriented databases -- such as the one built by Stonebraker's latest start-up, Andover, Mass.-based Vertica Systems Inc. -- store data vertically in table columns rather than in successive rows.

http://www.computerworld.com/action/article.do?command=viewArticleBasic&articleId=9034619&source=NLT_ES&nlid=42