CUBE DATA INTEGRATION ON RELATIONAL OLAP USING UNIFIED CUBE MODEL

• Main Author: Hidayat, Teddy
• Format: Theses
• Language: Indonesia
• Online Access: https://digilib.itb.ac.id/gdl/view/46908
• Institution: Institut Teknologi Bandung

OLAP is designed to assist users in analyzing a subject, for example sales in a company. The data is stored in a multidimensional form called the cube. With the increasing number of external data sources that can be utilized for business, cube does not only contain data that comes from one source. Ravat et al. (2016) conducted a research on how to integrate cube data in the form of Relational OLAP (ROLAP) and Web OLAP (WOLAP) into a form of unified view modeling called the unified cube. This thesis aims to apply the unified cube model to integrate data cube on ROLAP. In this thesis, the integration of data cubes in ROLAP with unified cube modeling is applied to data originating from external sources in the form of csv file containing measure and time series attributes. The main problem that must be addressed is the process of mapping data on extrinsic links, namely links that describe the relationships between dimensions of different data cubes. In this thesis, extrinsic links are implemented in the form of extrinsic link tables. In addition, several SQL templates have been developed to query an unified cube that is implemented in ROLAP. Tests have been carried out involving two data sources that form two data cubes in ROLAP. The tests show that the unified cube model applied in this thesis can be used to perform OLAP operations on two cube at the same time and on two measures in different cubes.