File:Maintaining Multimedia Data in a Geospatial Database (IA maintainingmulti1094517318).pdf

The maintenance and organization of data in any profession, government or commercial, is becoming increasingly more challenging. Adding components, whether those components are two- or three dimensional, further increases the complexity of databases. It is harder to determine which database software to choose to meet the needs of the organization. This thesis evaluates the performance of two databases as spatial functions are executed on columns containing spatial data using benchmark testing. Evaluating the performance of spatial databases makes it possible to identify performance issues with spatial queries. The process of conducting a performance evaluation of multiple databases, in this thesis, focuses on the measurement of each elapsed time within each database. The work already implemented in evaluating the performance of spatial databases did not explore a databases performance as it returned large and small result sets. The overhead of returning large or small result sets was not considered. Therefore, a custom test was developed to engage the aspects of prior work found beneficial. Using a database the researchers built with well over one million records, the elapsed time in adding records was measured. The elapsed time of the spatial functions queries was measured next. The results showed areas where each database excelled given multiple conditions. A different look at PostgreSQL and MySQL as spatial databases was offered. Given their results, as each database produced result sets from zero to 100,000, it was learned that the performance of each database could differ depending on the volume of information it is expected to return.