Posted in Information Technology & Systems, Total Reads: 748
De-normalization is typically used in relational database where it is an approach to enhance the performance of the database. This involves the process of speeding up the data retrieval or read activity by inserting redundant data to normalized database in order to remove the problems associated with database queries which combine the data in different tables into one single table. De-normalized database is however different from the one which is not normalized.
E.g.: An array of global data variables of say, student names is tied together with a single purchase of a book in the purchase history. In this case, the primary aim of de normalization would be to create a separate table or structure so that queries on the book of purchase will not affect the information tied onto it i.e. the student account. Hence, a database administrator will ensure that this information is separated with redundant data by database handler in order to facilitate the ease of working separately on one component of information. Here again, the read performance is improved by the specific way the database handler searches for a particular item. Difficulties in denormalization include careful documentation due to the possibility of recording anomalies resulting from the addition of redundant data in the database.