Faculty of Economics
Collecting raw data from multiple sources presents various challenges, such as dirty data, sensitive data, lack of a single version of truth (SVoT), incompatible data formats and massive data loads, all of which may potentially compromise data reliability and lead to discrepancies and errors.
As a number of source systems grows, so does the data volume. To avoid redundancies, organisations seek advanced enterprise warehouse solutions designed to effectively process large sets of data, optimize database queries and guarantee interoperability between datasets.
Our data integration architecture is based on the Data Vault model that encompasses the best aspects of both the third normal form and the star schema and provides a uniquely-linked set of normalized tables that keeps track of historical data. We further enhanced its benefits by using ontology modelling to ensure semantic integration.
The three foundational entities of the Data Vault model are hubs, links and satellites: hubs represent core business concepts through a list of business keys; links manage the relationships between keys; satellites store the descriptive attributes related to hubs and links. One of the distinct features of the Data Vault 2.0 is the hash function applied to standardize the keys that come in various formats.
To learn more about the Data Vault model, read the Data Vault Handbook.
Sources:
The Data Vault Handbook (2025).
Linstedt D., Olschimke M. (2016). Building a Scalable Data Warehouse with Data Vault 2.0. Morgan Kaufmann.
Russian Federation, 119991, Moscow, GSP-1, Leninskie Gory,
Lomonosov Moscow State University,
Building 1, Building 46 (New Academic Building 3), Faculty of Economics, rooms 546, 548, 550
Department of Economic Informatics
Our website is at econ.msu.ru
+7 (495) 939-30-67 — Secretary
+7 (495) 939-57-25 — Faculty Office
Email: ecinf.econ@org.msu.ru