Shulou(Shulou.com)06/03 Report--

Big data service model design: unknown wife

Li Fudong

Li Fudong

Big data, operator, architect, trainer

This article is from Li Fudong's "big data Operation-New thinking of Service-oriented Enterprise Architecture" 3.4

Editor's note: actions are done in thinking and destroyed by others. The operation-oriented data model focuses on the support of "rows", while the analysis-oriented data model focuses on the support of "thinking".

In order to see the whole process of data formation, aggregation, integration and use, we design it from two stages: the operation-oriented data model and the analysis-oriented data model.

The transformation from an operation-oriented data model to a decision-oriented data model is shown in figure 3-4-1:

Figure 3-4-1 Transformation of data from an operating environment to an analysis-oriented environment

As can be seen from figure 3-4-1, operational data comes from transactional applications, which will be stored in databases supporting transactional applications, and data from different sources will be aggregated to the data warehouse, and then data marts for different topics will be formed through tools and means such as ETL.

Take telecom operators as an example, in operation-oriented transactional applications, customers and their ordering data will be recorded. If the customer's business usage data is added, a customer-centered, 360-degree data set can be formed. Through the classification and aggregation of data, it can help enterprises to comprehensively see the whole process of customers from consultation, ordering, payment, use, declaration, complaint and suggestion. In order to achieve the purpose of helping enterprises make decisions.

According to the purpose of data service, the data model is divided into operation-oriented data model and analysis-oriented data model. The operation-oriented data model mainly supports enterprises to complete the operations of data addition, deletion, modification and query, and helps enterprises to complete the tasks of construction, production, operation and internal management. The analysis-oriented data model mainly supports the statistical analysis functions of different topics and different dimensions.

It can be seen that the operation-oriented data model focuses on the support of "row", while the analysis-oriented data model focuses on the support of "thinking". "doing is done and destroyed by following", without action, there is no material for thinking, without deep thinking, the action is likely to deviate from the direction, and enterprises need to unify "action" and "thinking".

From the support level of the data model to the enterprise application, the data model is divided into strategic model, tactical model and operational model, as shown in figure 3-4-2:

Figure 3-4-2 data model at three levels of strategy, tactics and operation

At the strategic level, the analysis-oriented data model supports senior managers to complete the strategic work, including factory location, channel location, merger and acquisition planning, investment and financing planning, environmental impact analysis, unconventional capital budget and so on.

At the tactical level, the analysis-oriented data model supports the middle-level cadres to complete the management work, including marketing plan, sales plan, customer service plan, human resource plan, financial budget and so on.

At the operational level, the analysis-oriented data model serves the daily production and operation activities of enterprises, including personal loan credit, production schedule, inventory control, maintenance plan, quality control and so on.

The operation-oriented data model and the analysis-oriented data model are like the relationship between automobile engine and throttle / brake. The operation-oriented data model is like an automobile engine, which ensures normal operation and turning, while the analysis-oriented data model is like a driver who determines to step on the accelerator or brake according to road conditions. The two must cooperate to ensure that the business activities of the enterprise are carried out in an orderly manner.

(this article is excerpted from: Li Fudong's "big data Operation-New thinking of Service-oriented Enterprise Architecture" 3.4, forwarded by the author, the picture is from the network)

Posted at 15:05

Welcome to subscribe "Shulou Technology Information " to get latest news, interesting things and hot topics in the IT industry, and controls the hottest and latest Internet news, technology news and IT industry trends.