Cost Optimization in Azure Data Engineering
Montazeri, Kiana (2025)
2025
All rights reserved. This publication is copyrighted. You may download, display and print it for Your own personal use. Commercial use is prohibited.
Julkaisun pysyvä osoite on
https://urn.fi/URN:NBN:fi:amk-2025112830539
https://urn.fi/URN:NBN:fi:amk-2025112830539
Tiivistelmä
This bachelor thesis discusses saving costs in data engineering in Azure. Specially the services which are more expensive in this process. Most of the companies are moving their data systems to cloud, as it's more flexible and scalable. Cloud costs can grow very quickly if they are not watched carefully. The main goal for this thesis is to reduce costs while it stays efficient and reliable.
This study is a product-based project, meaning it was done to solve a real problem for the author's employer, Eltel Networks Oy. Eltel builds and keeps up power and communication networks The Company is moved from the old on-premise data platform to Azure, Eltel is planning to move other countries' (like Norway and Sweden) data to cloud in near future, so the amount of data will increase and that's why this research will be important as it will be used for proper planning for this new Project to optimize costs when they grow after this change. services like Azure Synapse Analytics, Azure Data Factory (ADF), and Azure storage are much more expensive and should be considered the most.
The thesis first explains the basic concepts related to data engineering, such as the Data Vault methodology (using Hubs, Links, and Satellites), and the difference between ELT and ETL. It then talks about the biggest cost drivers in Eltel's system: Azure Synapse Analytics (especially the Dedicated SQL Pool), Azure Data Factory, and Azure Blob Storage/Gen2. To manage these costs, the thesis studies Azure’s built-in tools like the Azure Pricing Calculator and Azure Advisor. It also introduces FinOps (Financial Operations), which is a culture that helps finance, engi-neering, and operations teams work together to manage the expenses.
The research approach involved analyzing Azure's pricing rules and using Eltel’s actual cost data. The main product created was an Azure Cost Analysis report and clear documentation for the company's BI team. The report focused on Synapse, ADF, and Storage for the month of October 2025 and confirmed they are the highest cost services.
The biggest cost savings can be achieved by optimizing the major compute and data movement services. The main strategy involves matching resource usage to actual work and taking advantage of built-in Azure features for better rates.
In the long run, the thesis recommends that Eltel use consistent tagging on all resources and routinely follows the advice from the Azure Advisor to ensure money is always being saved and waste is avoided. This thesis helps developers and companies understand how to manage Azure data engineering costs better and supports long-term cloud cost strategies.
This study is a product-based project, meaning it was done to solve a real problem for the author's employer, Eltel Networks Oy. Eltel builds and keeps up power and communication networks The Company is moved from the old on-premise data platform to Azure, Eltel is planning to move other countries' (like Norway and Sweden) data to cloud in near future, so the amount of data will increase and that's why this research will be important as it will be used for proper planning for this new Project to optimize costs when they grow after this change. services like Azure Synapse Analytics, Azure Data Factory (ADF), and Azure storage are much more expensive and should be considered the most.
The thesis first explains the basic concepts related to data engineering, such as the Data Vault methodology (using Hubs, Links, and Satellites), and the difference between ELT and ETL. It then talks about the biggest cost drivers in Eltel's system: Azure Synapse Analytics (especially the Dedicated SQL Pool), Azure Data Factory, and Azure Blob Storage/Gen2. To manage these costs, the thesis studies Azure’s built-in tools like the Azure Pricing Calculator and Azure Advisor. It also introduces FinOps (Financial Operations), which is a culture that helps finance, engi-neering, and operations teams work together to manage the expenses.
The research approach involved analyzing Azure's pricing rules and using Eltel’s actual cost data. The main product created was an Azure Cost Analysis report and clear documentation for the company's BI team. The report focused on Synapse, ADF, and Storage for the month of October 2025 and confirmed they are the highest cost services.
The biggest cost savings can be achieved by optimizing the major compute and data movement services. The main strategy involves matching resource usage to actual work and taking advantage of built-in Azure features for better rates.
In the long run, the thesis recommends that Eltel use consistent tagging on all resources and routinely follows the advice from the Azure Advisor to ensure money is always being saved and waste is avoided. This thesis helps developers and companies understand how to manage Azure data engineering costs better and supports long-term cloud cost strategies.