Cloud-based data warehouse

Challenge

A client in the telecommunications industry decided to build a new data warehouse based on a cloud environment. The focus was on address points for FTTH/FTTB fiber optic installations, covering approximately one million residential units (including both existing and new locations). The main goals for the new data warehouse were:

To gather dimensions and attributes for address points in the warehouse to meet current reporting needs
To reduce data processing and storage costs
To apply standards and best practices in data warehouse construction
To facilitate the integration of address point data with data from other areas, such as the orders area

Solution

Based on Sorigo’s experience in the telecommunications field and in building data warehouse solutions, our team designed and implemented a data warehouse for the client. The following activities contributed to its execution:

Developing and utilizing a standard for building data warehouse areas using the Data Vault 2.0 methodology and the Google BigQuery platform
Conducting a series of workshops and meetings with business stakeholders to define the relationships of dimensions and attributes (describing address points) in use cases and business processes
Preparing a comprehensive dependency analysis using Sparx Enterprise Architect, resulting in dependency matrices and relationship views focused on specific reporting dimensions
Implementing processes for loading address point data into the warehouse
Building a mechanism for managing subscriptions to specific dimension attributes describing the address points of fiber optic service installations
Providing an information layer and data mart for address point data to agency network teams (in collaboration with operators) and back-office consultants

Sorigo also recommended changes to systems based on the analysis and reporting requirements.

Result

Thanks to the implementation of a data warehouse based on the Google BigQuery platform and the Data Vault 2.0 model, the client:

Increased the efficiency of using address point data for business purposes
Reduced data processing and storage costs
Significantly improved data processing performance
Standardized the data model and solution architecture
Gained the ability to track attribute changes over time (change history), which is crucial for back-office consultants
Lowered the costs of introducing new attributes in defined dimensions
Made it easier to integrate address point data with data from other areas
Enhanced the solution’s scalability, flexibility, and security

Increased business process efficiency

Security

Performance

Flexibility