Abstract:
Database management systems have become the most important process since millions and billions of data transactions taking place every second. It comes as surprise that d...Show MoreMetadata
Abstract:
Database management systems have become the most important process since millions and billions of data transactions taking place every second. It comes as surprise that database optimization and tuning has become the main key research. If the database processes are not handled properly, it will lead to a slow system and it might cause a lot of errors and there is a possibility for the system to crash. Since database main task is storing and accessing the data according to the user needs through SQL operations, therefore there is a need to optimize the database operations by reducing their response times. There are many ways to optimize database operations, but among those; database tuning seems the most challenging area. There are many studies done on the improvement of database tuning approach however they are still suffer from a slow query processing time. Factors causing slow processing time normally are due to small-shared pool size and improper execution plans used during queries. This study focused on improving database indexing to overcome these two factors. A combination of clustered index, non-clustered index and bitmap is proposed as an integrated approach in query processing. These combinations of indexes are tested using complex and simple queries. The results of this experiment are being compared with the result from the existing indexing approaches when using the same datasets. The result shows that these combinations are able to optimize the database systems. The proposed solution able to provide a database system that is fast in data retrieval and an improved performance percentage of 10% to 20% depending on the query, and the dataset used. Indirectly, this proposed solution enables companies' database systems to achieve its highest potential with maximum performance.
Date of Conference: 12-13 October 2021
Date Added to IEEE Xplore: 25 November 2021
ISBN Information: