Jim Lawrence
accessd at shaw.ca
Sat Apr 3 04:31:47 CDT 2010
As database data becomes larger and more fragmented in it basic structure (if you want to call it a structure) here is a link to one of the new super NOSQL DBs used by many companies to manage this massive data. http://cassandra.apache.org This type of database does not use indexes but rather nodes which can be described as object node (groups of related data). There are three things about it may scare many away from investigating further. It is Linux (Debian) based, Java coded band will require some manual labour, at command prompt and a paradigm shift in concept from the traditional RDBMS but for those brave adventures and those with huge complex sets of data and want to be able to retrieve data at incredible speeds, this is Royal Jelly. (This is the data engine behind such data giants as FaceBook.) Once you get by the above reality, it does have a nice graphic interface, much like MS SQL There is no difficulty in getting started with the product as it is OpenSourced and very stable... field tested extensively. The data model is almost unlimited in its potential size, complexity and scalability. So why would anyone switch when they are comfortable or at least resigned to their relation database? Here is a link on why people with huge data sets are switching (The article is not on Cassandra but a contemporary and very similar product called APDB but it will give you a very good overview of the new genre of data storage products): http://thedailywtf.com/Articles/Announcing-APDB-The-Worlds-Fastest-Database. aspx Jim