Analytical Infrastructure Under Resource Constraints: An Integrated Mixed-Methods Assessment of Organizational Requirements and Database Architecture Performance
(2026) SYSK16 20261Department of Informatics
- Abstract
- This thesis examines the suitability of analytical infrastructure for decision support in organizations with limited dedicated analytics capacity. The study uses an integrated mixed-methods
design combining semi-structured interviews with a controlled TPC-H benchmark. The interviews identify organizational requirements, infrastructure and data flows, resource constraints, and priorities shaping architectural suitability. The benchmark evaluates analytical
performance across three database configurations: a normalized OLTP schema with columnstore indexing, a dimensional OLAP schema with columnstore indexing, and a dimensional OLAP schema with rowstore indexing.
The findings show that architectural suitability depends on technical... (More) - This thesis examines the suitability of analytical infrastructure for decision support in organizations with limited dedicated analytics capacity. The study uses an integrated mixed-methods
design combining semi-structured interviews with a controlled TPC-H benchmark. The interviews identify organizational requirements, infrastructure and data flows, resource constraints, and priorities shaping architectural suitability. The benchmark evaluates analytical
performance across three database configurations: a normalized OLTP schema with columnstore indexing, a dimensional OLAP schema with columnstore indexing, and a dimensional OLAP schema with rowstore indexing.
The findings show that architectural suitability depends on technical performance alongside
factors such as competence, data movement, coordination, maintainability, and analytical requirements. The benchmark shows that the columnstore-indexed OLTP configuration performed comparably to the dimensional columnstore configuration for many analytical queries,
while some join-intensive workloads favoured the dimensional schema.
By interpreting benchmark performance in relation to the organizational requirements identified through the interviews, the study shows that integrated HTAP-oriented architectures may
be suitable under some conditions, while separated OLAP architectures remain advantageous
where workload or organizational requirements justify them. The study contributes to Information Systems research by demonstrating how organizational requirements, resource constraints, and technical performance can be considered together when assessing analytical infrastructure for decision support. (Less)
Please use this url to cite or link to this publication:
https://lup.lub.lu.se/student-papers/record/9250039
- author
- Magnell, Jack LU ; Törnqvist, Nils LU and Bergman, Felix LU
- supervisor
- organization
- course
- SYSK16 20261
- year
- 2026
- type
- M2 - Bachelor Degree
- subject
- keywords
- Analytical Infrastructure, Resource Constraints, Decision Support, Mixed Methods, HTAP, OLTP, OLAP, Columnstore Indexing, TPC-H
- language
- English
- id
- 9250039
- date added to LUP
- 2026-09-03 15:36:47
- date last changed
- 2026-09-04 13:43:16
@misc{9250039,
abstract = {{This thesis examines the suitability of analytical infrastructure for decision support in organizations with limited dedicated analytics capacity. The study uses an integrated mixed-methods
design combining semi-structured interviews with a controlled TPC-H benchmark. The interviews identify organizational requirements, infrastructure and data flows, resource constraints, and priorities shaping architectural suitability. The benchmark evaluates analytical
performance across three database configurations: a normalized OLTP schema with columnstore indexing, a dimensional OLAP schema with columnstore indexing, and a dimensional OLAP schema with rowstore indexing.
The findings show that architectural suitability depends on technical performance alongside
factors such as competence, data movement, coordination, maintainability, and analytical requirements. The benchmark shows that the columnstore-indexed OLTP configuration performed comparably to the dimensional columnstore configuration for many analytical queries,
while some join-intensive workloads favoured the dimensional schema.
By interpreting benchmark performance in relation to the organizational requirements identified through the interviews, the study shows that integrated HTAP-oriented architectures may
be suitable under some conditions, while separated OLAP architectures remain advantageous
where workload or organizational requirements justify them. The study contributes to Information Systems research by demonstrating how organizational requirements, resource constraints, and technical performance can be considered together when assessing analytical infrastructure for decision support.}},
author = {{Magnell, Jack and Törnqvist, Nils and Bergman, Felix}},
language = {{eng}},
note = {{Student Paper}},
title = {{Analytical Infrastructure Under Resource Constraints: An Integrated Mixed-Methods Assessment of Organizational Requirements and Database Architecture Performance}},
year = {{2026}},
}