Evaluating ERD Models and RAID-Based Storage for Query Performance Optimization in Relational Databases

Abstract

The amount of data stored in magnetic disks (e.g., floppy disks) increases by 100% each year for each department in a company, necessitating efforts to maintain an optimal database system. Designing a database is the initial step in creating a system with optimal performance. However, database design alone is not sufficient to enhance performance. One approach to improving data transaction speed is by optimizing query processing. This research evaluates different relational database models using varying amounts of data. Query costs are analyzed using the Cost-Based Optimizer method and access time measurements. The results of this study provide insights for database administrators in designing relational database models effectively and selecting appropriate query structures to optimize database performance. The findings indicate that: (1) database design can be optimized by separating entities based on specialized usage, and (2) factors such as record count, attribute size, query type, use of unique or primary keys, order-by clauses, index sequences, and SQL function usage significantly impact query cost and overall performance.

Keywords

• database design
• query optimization
• relational database model
• cost-based optimizer
• query performance