Evidence record · UDINUS · SDGs
- Unit
- Rektorat UDINUS
- Period
- 2018
- Goals
- Record
- 99b7c59a-03aa-4d37-9972-097b1e90db28
- Published
- 2026-10-30
Universitas Dian Nuswantoro conducted research on query optimization in a client/server environment of a distributed database using health epidemiology data based on dengue fever data and Geographic Information Systems. The activity addressed the need for an appropriate method for query processing in a distributed database. Query processing requires attention because a cost-based query process may involve a number of attributes and visited sites.
The research examined the separation of attributes that are not required by a query. A query operation may search for data from various attributes in a scattered database table even though the process does not require all table attributes. The study considered minimum query operating costs, including communication costs and access fees. It examined vertical fragmentation as a method for separating attributes in a health database table. The researchers compared the Bond Energy Algorithm and the Graphic Based Vertical Partitioning algorithm.
For both algorithms, the initial result of vertical fragmentation was the determination of the types of attributes separated from a number of specific query processes. The separated attributes from each algorithm were compared and evaluated using Partitioned Evaluator to achieve the access cost of several attributes. The results show that the Graphic Based Vertical Partitioning algorithm was more optimal for vertical table fragmentation applied to query operations on a distributed dengue fever database in the health field. Compared with the Bond Energy Algorithm, it had less computational complexity, produced a higher Partitioned Evaluator value, and had lower query execution time.