Description
Data, analytics, and artificial intelligence (AI) play a critical role in modern business operations.In 2023, 41% of Finnish companies engaged in data analytics; and by 2024, 24% were utilizing
AI. In the 2020s, cloud-based data warehousing platforms emerged, prompting companies to
transition from traditional on-premise solutions to more flexible cloud offerings. These platforms
enable rapid deployment of advanced features such as real-time streaming, data governance,
self-service analytics, and integrated machine learning (ML) and AI capabilities. However, despite
these technological advancements, effectively modeling and managing data in a structured and
systematic manner remains crucial for extracting meaningful insights from these platforms. This
thesis develops a comprehensive framework for designing cloud-based data warehouses, lever-
aging the Data Vault 2.0 (DV 2.0) methodology to address the challenges of scalability, flexibility,
and efficiency in modern data systems.
The research examines three primary data modeling methodologies: normalized, dimensional,
and DV 2.0, with a particular focus on DV 2.0's application in cloud-based environments. It ex-
plores how DV 2.0 differs from other methodologies and highlights the core principles of data
modeling, emphasizing their importance. The study includes a literature review of data modeling
concepts, followed by empirical findings from semi-structured interviews with industry experts.
These interviews cover key themes such as architectural approaches, data modeling methodolo-
gies, new analysis requirements, change management, auditability, and best practices.
The interviews identified several challenges in data warehouse implementations, categorized into
technical and business aspects. Technically, inconsistencies in modeling, lack of guidelines, and
poor development practices were major obstacles. From a business perspective, unclear require-
ments, insufficient ownership, and limited resources were key concerns. Interviewees stressed
the importance of aligning data warehouse implementation with organizational resources and
scope, avoiding unnecessary complexity while maintaining clear guidelines. They also highlighted
the need for agile project management with structured planning and defining roles and ownership
for long-term success. Choosing the right technology, with consideration for scalability, cost-ef-
fectiveness, and automation, alongside strong governance and best practices, is critical for man-
aging system complexity over time.
This research provides insights into the benefits and limitations of DV 2.0, particularly in cloud-
based data warehouses, and proposes a framework that integrates DV 2.0's principles within
modern data lakehouse platforms. The main findings suggest that a data warehouse should follow
a multi-layered approach with clear modeling guidelines. The thesis concludes with a discussion
on the practical implementation of the proposed framework and offers recommendations for future
research, including empirical validation and exploration of alternative modeling methodologies.
| Period | 21 Mar 2025 |
|---|---|
| Examinee | |
| Degree of Recognition | International |