Data Visualization Systems Designation

This part of the content is intended to briefly introduce the design ideas of data visualization products such as enterprise BI systems or big data analysis platforms, provide reference for product designers, and analyze and decrypt such data products for interested readers.

1. Data source management

Such products currently support more and more data sources. After all, data is the basis for data analysis and visualization. It is impossible to connect to the data source, and all the following work will not be carried out.


When designing such products, it is first necessary to complete the investigation of the company's existing data sources, and then select the data source to be supported by the project according to the actual scenario.

Because different databases have different advantages and disadvantages, in addition to accurate data and reasonable interaction of styles, visual reports also have requirements for data query speed. We cannot allow users to wait for a few minutes after performing a click operation.

For example, when the amount of data reaches 100 million, MySQL is not suitable. At this time, it is more suitable to choose some distributed databases, such as HBase.

According to the amount of data on the platform and the company's existing capabilities, the selection of data sources to be supported is the first step in the design of such products.

2. Chart library management

The chart library is like an "ammunition library", but the packaging of chart components is a protracted battle, which is time-consuming and labor-intensive, and requires a lot of front-end development resources. However, we can use open source chart libraries such as ECharts:

According to the needs of visualization, constantly improve the chart library of the self-developed system, which allows data analysts to use the visualization tools to achieve "comfortable", then it is considered a little accomplished.

3. Report development and distribution

Report development and distribution is the core function of the data visualization system. While implementing report development, this part also needs to complete the configuration of permissions (view permissions and data permissions).

To achieve this part of functional design, there is a "shortcut" to learn how to implement mainstream visualization tools on the market. Their detailed product help documents are the "source of inspiration" for self-developed product design.

As for the control of permissions, there are many ways, such as single-point reading user permission table, etc., the control of data permissions is a must for data security.

Conclusion

Data visualization needs to be user-centric and based on user experience to output visual reports to empower business.