Issue link: https://epubs.iltanet.org/i/98946
Evolving Trends in Database Technology Four Types of NoSQL NoSQL databases can be categorized as one of the following: Key/Value pairs feature ultrafast lookups of a payload (data) based on a key that the user or application supplies. A typical use includes content caching. One radical departure is that this type of NoSQL typically does not have a schema, but this can be seen as an advantage. Examples include Couchbase server (formerly Membase), Redis and Microsoft Azure Table Storage. Document databases are nonrelational storage systems that feature optional schema on-the-fly. Variable length records are the norm. The most easily visualized use case is that of a personal contact record stored in an array. There are a lot of fields that may or may not be used. Examples are MongoDB and CouchDB. Column store databases turn a relational database on its side, with an optional schema and deliberate denormalization. They are very efficient at aggregations and make great highly scalable indexes for data stored in multiple places. Because they are so much more efficient than traditional row stores, performance gains of up to a factor of 50 can be realized. Examples include Cassandra, Bigtable and Hbase. Graph databases are very different from any other database you have ever seen. Its closest approximation is to that of the old network databases that predate the relational model. Graph databases are very good at representing many to many relationships; something that has always been difficult to do well in the relational model. The most common use case is to model a roadmap or a social network. An example is Neo4J. NoSQL world are Datastax (Cassandra), 10Gen (MongoDB) and Couchbase (Couchbase Server). In the Hadoop space, Cloudera and Hortonworks (among others), offer distributions for download. In most cases, the company offers additional value or functionality for licensed customers. Not If, but When Nonrelational databases are rapidly moving onto the radar of C-level executives. It's best to research them and develop a strategy on how they can fit into your organization. It's not a question of if you will adopt NoSQL technology in the enterprise, but when. FUN FACT Some people have heard of Hadoop in the media and think it's a NoSQL database. Hadoop is actually a distributed framework for processing and performing analytics on unstructured data. It contains a NoSQL database, Hbase, along with about a dozen other pieces of open-source software bundled in a distribution, much like Linux. Thinking About NoSQL? If you are interested in considering a NoSQL database, it's advisable to take a look at some of the enterprise partners who can help deliver a solution. In many of the opensource business models, a service provider invests in the technology heavily and can deliver support, training and implementation services. Some of the leaders in the ILTA White Paper 39