The difference between the data warehouse and the transaction type database

The specific database that we create for a program is generally called a transactional database (Transaction), and its data Warehouse (DW) is very different.

1. Transactional databases help people perform activities, and data warehouses help people plan. For example, a transactional database might show which seats on a flight are empty, so passengers can make reservations. The Data warehouse is used to show the historical information of the vacancy rate, so that the flight manager decides whether or not to schedule the navigation in the futureclass.
2. The transactional database focuses on the details, and the Data Warehouse focuses on high-level aggregation. For example, an adult only wants to buy the most popular children's books without caring about itsinventory situation. But the managers in charge of books on shelves also care about the sales and trends of books. The different implied differences between this are the numberData in a warehouse is usually a numeric type that can be used for summarization.
3. Transactional databases are typically designed for specific programs, and data warehouses are used to consolidate data from different sources. For example, the order handler--the database will containdiscount information for each order, but does not include cost overruns for the product. The corresponding product management program will contain detailed cost information but will not contain salessales Discount. But you can combine the two messages into a single data warehouse, and you can calculate the actual sales profit of the product.
4. The transactional database focuses on the present, while the Data Warehouse focuses on history. For example, a bank account, each transaction-that is, each time the deposit and withdrawal will changethe value of the account balance, but the transaction system rarely maintains the historical balance. But in the data warehouse, you can store many years of transactional data (which may be aggregated), the balance snapshot is also stored. This allows you to compare the current value to the history. When making a decision, you can view historical trends.
5. Transactional databases are mutable, which is easy to understand. But the data warehouse is stable, and its information is updated at regular intervals (possibly monthly,per week or hour), and ideally, the update will only add new values for the new time period, without altering the previously stored values.
6. The transactional database must provide quick access and updates to the details, and the data warehouse must be quick to obtain or update the high summary information. So the businessthe design optimization of the database and the design optimization of the data warehouse may be different. Also, querying a transactional database for administrative purposes generates reports that can cause thetransaction program run performance degradation

The difference between the data warehouse and the transaction type database