EFFICIENT LIKE COMMANDS FOR DYNAMIC DATA RETRIEVAL AND REPORT GENERATION FROM PARALLEL DATABASES.

Abstract

Data partitioning is a typical feature in parallel database systems. Data partitioning splits a table into smaller parts which can be accessed, stored and maintained independent of one another. In order to improve the query performance and on the whole manageability of the database system partition techniques are used. Data partitioning simplifies administrative tasks like data loading, removal, backup, statistical maintenance and storage provisioning. The most common task performed in all the databases is search or retrieval of data. Retrieval of data consumes time based on the database size. If the database is of very large size then data retrieval consumes more time and if the database is of small size then data retrieval takes less time. The project deals with efficient data retrieval from normal database and parallel databases dynamically. In enhanced search, search can be done based on any attribute dynamically. The search result is produced in a table format. From the resulting table any of the records details can be printed in a report format. For the retrieval of data three different like queries are used. '".$key_word."' Like query is used for numerical attributes and the '".$key_word."%' , '%".$key_word."%' are used for non numeric attributes. The precision of the above like commands are analyzed. The proposed techniques improve search performance and reduce the data retrieval time.