- Not every institution doesn’t have the same needs for data, so they need to have different data warehouse architecture
- Data warehouse came from different sources → semi-structured data or structured data
- The purpose of the data warehouse itself is to maintain the single truth of sources and also to easily visualize historical data and avoid data inconsistency
- the biggest problem of scalable enterprise: is the fragmentation of the database in a vertical way (each operational team has different needs for the database)
- Why do we need to centralize into a single data warehouse?
- Aggregation and collection of all the data
- Give an integration of insights
- Supporting decision making
- And also resources sharing for each division
- 4 characteristics of data warehouse
- Subject-oriented
- Integrated
- Time-variant
- Non-volatile