Cubes ROLAP sous forme de SDDS multidimensionnelles

Abstract

Scalable Distributed Data Structures (SDDS) is a data model specifically designed for distributed environments, where data must be dynamically partitioned and distributed across multiple servers while ensuring system scalability. An SDDS file consists of records that are dynamically distributed over servers. These records are organized based on various types of SDDS, each relying on specific algorithms, such as Linear Hashing (LH*), Range Partitioning (RP*), or Digital Hashing (CTH*). A key feature of SDDS is the elimination of any centralized address component, thereby simplifying communication between clients and servers. By removing the need for a central coordination point, which can act as a bottleneck, this architecture significantly reduces the number of exchanged messages and shortens data access times, ultimately improving the overall efficiency of distributed systems. Meanwhile, the Data Warehouse (DW) plays a critical role in decision support systems by enabling the analysis and exploration of large volumes of multidimensional data. These data are represented in the form of a cube and utilized through Relational Online Analytical Processing (ROLAP), which leverages relational databases to store and query the data effectively.