Big Transactional Data Set?
No Problem!
Project: SSAS Tabular Model to provision reporting for Retail Gaming Giant’s Transactional Big Data
Type of data: Transactional bet information by date, shop, product, event, customer etc.
- Size of Model: 1TB
- Number of users: 15,000
- Number of shops: 3,500
- Platform: On-Premises SQL Server Analysis Services Tabular Model and Power BI (Direct Query mode)
Highly performant and optimised model against big data
Reports dynamically run and display the data in less than 2 seconds.
Full Deep-Dive Capabilities
Ability to drill down summary information by key variables to interactively identify what is driving performance accross the business.
Consolidated Fact Table
Combines Data from Multiple Sources into a Single BET Table
- Retail Sports Bet Transactions
- Retail Self-Service Terminal Transactions
- Retail Gaming Terminal Transactions
- Online Digital Transactions
Aggregation Layers
4 Fact Tables of Different Grain, to provision queries efficiently
- 2 Year BET Table (2 Billion Rows)
- 2 Year Summary Table (48 Million Rows)
- 7 day BET Table (19 Million Rows)
- 7 Day Summary Table (460K Rows)
Built in Business Logic
A single source of truth and presentation layer for analysts and onward consumers of data.
Built in Time-Intelligence
The ability to filter measures like Sales by time-frame without needing to write complete date logic.
Switchable Facts, but a singular measure with switches
A single measure for sales, but ability to traverse between reporting against summary and more detailed Fact tables depending on performance and reporting needs.
