Lecture
The basis of its classification is the sectoral principle of the application of HD.
Билл Инмон
Financial Data Warehouses
In most cases, financial HD organizations are built first. Creating financial HD is a necessary component of the financial infrastructure of any organization.
Financial data is always the focus of management. Therefore, it is very easy to attract interest in creating such an information system.
The financial activity of most organizations (with the exception of financial institutions) is small, so the volume of financial data is not very large, the rate of data receipt is also small. Financial data is well structured. Therefore, the available software and hardware allows you to create and maintain compact financial CD.
Finance covers all aspects of the functioning of the organization and have one common denominator - money.
Financial data, by its nature, has a structure that is directly affected by the daily practice of financial information processing.
For these reasons, finance is becoming the most preferred area for building corporate HD.
Insurance Data Warehouses
HD in the field of insurance, with some minor exceptions, are similar to all others. The first exception (characteristic of Western companies) is that the duration of the existence of existing HD is very long. These CDs contain data that is very old (before the beginning of the 20th century)
The second difference between these CDs is determined by the dates, information about which is stored in this field of activity. The insurance environment, for whatever reason, is characterized by the presence of a huge number of business-related dates, more than in any other activity. The third difference is that these HDs use their business cycle. Most organizations have a very limited and short economic cycle.
Human Resources Data Warehouses
HD for human resources management are very significant differences from other HD. The first difference is the number of subject areas. Such HD inevitably has one important subject area - this is an employee. Practically everything else is subordinated to this area or occupies a secondary position.
The main difference with HD for human resources management is that they use very few transactions.
Global data storage
Global data warehouses are designed for a global view of the organization’s activities. There are three types of such HD:
- Geographically prevalent data processing.
- Functionally prevalent data processing.
- Sectoral prevailing data processing.
A feature of global CD is that there are often very few common dimensions at the global level. The only common dimension is money. And business integration can only be achieved with its help. Other dimensions may or may not make sense at the global level. Global HD should continuously respond to possible changes in business data. In this case, such changes, as a rule, are permanent. Therefore, the structure and technology used to host and maintain a global HD should allow these continuous changes to be maintained.
Data warehouses with new data discovery capabilities
CDs that support new data discovery technology are a hybrid of classic CDs. They are used to perform powerful statistical data processing. These HDs are:
- very detailed;
- deeply historical;
- optimized for statistical analysis.
In addition, for such HD is characterized by an orientation to a project. This means that, unlike all other types of CD, in most cases they are no longer used immediately upon completion of the analysis for which they were created.
Another important difference between CD and analysis capabilities is that it often includes external data. Such data is very useful in terms of predicting changes in business data, which is not so easy to see without their participation.
Telecom data warehouses
A distinctive feature of these CDs is that they are largely determined by data regarding the fact of telephone conversations. Of course, there are many other types of data in the telecommunications industry. But no other area of HD is predetermined to such an extent by the size of one subject area — the details at the conversation level.
Comments
To leave a comment
Databases, knowledge and data warehousing. Big data, DBMS and SQL and noSQL
Terms: Databases, knowledge and data warehousing. Big data, DBMS and SQL and noSQL