Clustering methods in data fragmentation 1
Adrian Sergiu Dărăbant, Laura Darabant · 2011
This paper proposes an enhanced version for three clustering algorithms: hierarchical, k-means and fuzzy c-means applied in horizontal object oriented data fragmentation. The main application is focusing in distributed object oriented database (OODB) fragmentation, but the method applicability is not limited to this research area. The proposed algorithms produce fragments for an OODB database based on the analysis of inter-class relationships and user queries (applications) running on the system. Each class extension is clustered and the quality of resulting fragments is then evaluated and compared between the proposed algorithms and with results obtained from other object-oriented fragmentation techniques. Numerical experiments on different databases show an average improvement in query processing time of 17-30%. The test scenarios take in account different database sizes. The methods are applied to a small, medium and large database in order to verify their scalability.