An evaluation of vertical class partitioning for query processing in object-oriented databases

Abstract

Abstract--Vertical partitioning is a design technique for reducing the number of disk accesses to execute a given set of queries by minimizing the number of irrelevant instance variables accessed. This is accomplished by grouping the frequently accessed instance variables as vertical class fragments. The complexity of object-oriented database models due to subclass hierarchy and class composition hierarchy complicates the definition and representation of vertical partitioning of the classes, which makes the problem of vertical partitioning in OODBs very challenging. In this paper, we develop a comprehensive analytical cost model for processing of queries on vertically partitioned OODB classes. A set of analytical evaluation results is presented to show the effect of vertical partitioning, and to study the trade-off between projection ratio vs. selectivity factor vis-a-vis sequential vs. index access. Furthermore, an empirical experimental prototype supporting vertical class partitioning has been implemented on a commercial OODB tool kit to validate our analytical cost model.

Vertical Fragmentation in Databases Using Data-Mining Technique

“…Because of the criticality of the database performance, several researchers have contributed enormously to vertical partitioning problem for over two decades. Database partitioning has been applied in centralized relational databases (Ceri, Navathe, & Wiederhold, 1983;Cornell & Yu, 1990;Hoffer & Severance, 1976;Ng, Gorla, Law, & Chan, 2003;Song & Gorla, 2000), distributed databases (Baiao et al, 2004;Cheng, Lee, & Wong, 2002;March & Rho, 1995;Ozsu & Valduriez, 1996;), Data Warehouse Design (Ezeife, 2001;Furtado, Lima, Pacitti, Valduriez, & Mattoso, 2005;Labio, Quass, & Adelberg, 1997), and Object-Oriented Database design (Fung, Karlapalem, & Li, 2002;Gorla, 2001). Hoffer and Severance (1976) consider the vertical partitioning problem by applying bond energy algorithm on similarity of attributes, which are based on access patterns of transactions.…”

“…With reference to object-oriented database design, Gorla (2001) used genetic algorithm to determine the instance variables that should be stored in each class/ subclass in a subclass hierarchy, so that the total cost of database operations is minimized. Fung et al (2002) analyze vertical partitioning of classes/ subclasses in object-oriented databases for class composition hierarchy and subclass hierarchy and develop the associated cost functions for query processing under the cases of large memory and small memory availability. Partitioning has been applied in data warehouses in order to optimize OLAP queries by designing parallel database clusters (Furtado et al, 2005) and to generate ROLAP data cubes based on optimized data partitioning technique (Chen, Dehne, Eavis, & Rau-Chaplin, 2006).…”

“…The amount of data transferred and the number of disk accesses have been the most commonly used methods of evaluating partitioning schemes by previous researchers (e.g., Cheng et al, 2002;Cornell & Yu, 1990;Fung et al, 2002;Gorla, 2001). In our research, we do not include cost of index access and storage cost.…”

Query Optimization on Distributed Database Dengue Fever by Minimizing Attribute Involvement

“…GBVP algorithm has less complexity in computing and produces meaning fragments with graphs. The affinity matrix is transformed into an affinity graph in order to partition the fragment based on defined rules and steps (Fung et al, 2002).…”