The large screen of data visualization has been widely applied to various business scenarios, such as group operation cockpit, intelligent operation center, emergency command center, security monitoring center, property monitoring center, transportation and logistics scheduling center, financial transaction center and so on.

Indeed, I am deeply touched. As an engineer (commonly known as my cousin) who has developed more than 300 statements, I have found more and more business needs about visualization in the past two years, especially for some cockpit and large data screen.

It takes nearly a month to make a large screen, and the technology is always teasing that white people like to “show off”. But from the point of view of the business and senior management, this kind of thematic, listing the core key indicators of the business, reflecting the dynamic in real time, is really a good way to monitor the business.

I used to make a couple of big data screens at a securities firm. Data will be collected from the source business system through the data warehouse, and the data will be cleaned and sorted out in the form of statements to generate statements and display. Key indicators such as turnover, trading volume, number of accounts opened and number of accounts eliminated will be counted to form a large screen of operation monitoring.

Because brokers and banks have a very important daily reconciliation, the daily transfer of funds into the amount of money, is very useful for monitoring the banking business, each time is not overtime, some customers said that the account rate is slow, this can be well monitored, and has done real-time monitoring.

The big screen is hung in the business department, and the leaders will know how the trading situation is today and how the marketing operation effect is when they come to see it, which is of great significance to the senior leaders.

So how to develop large data screens?

It’s all over the Internet, but anyone who can make a graph says they can make a big screen.

This is just talking about the front end, the actual development, where does the data come from? How to summarize, data caliber, definition of indicators, real-time refresh rate of data, which hardware devices to use, how to design the large screen, as well as performance and other back-end issues need to be considered. I’m a tech guy, technically speaking.

For conventional data visualization methods, we can choose to directly read the database, draw by drawing software/library, write HTML, CSS and JS codes in depth, introduce popular visualization such as Echarts and Hicharts in the process, and then design, adapt and finally form a self-built front-end display effect.

In addition, the pursuit of efficiency can choose mature visualization platforms, such as Ali DataV and FineReport, BI reporting suite. These suites often have a series of chart templates plus push-and-drop and visual configuration pages that allow us to quickly build large visual screens.

There is also a kind of outsourcing development customized, but the actual development is nothing more than the above technology.

If the report is developed internally or customized for Party A in batches, it is recommended to use a more efficient report tool, such as FineReport.

Reason 1: The front-end rendering effect can basically cover

The visual presentation effect of large screen is the face of the whole project, which is often referred to as a “master project” internally. Because those who watch or use large screen are often high-level decision makers of enterprises, the presentation effect and degree of individuation are naturally higher.

The large data display, essentially a cockpit page, consists of tables, graphics, controls, and other elements, each of which can be viewed as a separate component.

The visualization of the large screen is essentially a cockpit page, composed of tables, graphics, controls and other elements, which naturally leads to various open source and rich libraries on the market.

However, based on my actual project experience, most of the conventional business large screen uses common basic graphics, such as column and its column graph, line graph, pie chart and its composite graph. With maps, index cards, text components, dynamic table data rotation; In other specific model scenarios, such as the production shop, production models may use advanced 3D graphics, as well as some 3D rotation effects.

These report tools FineReport can basically meet the requirements, its own 50 or 60 chart styles, configured with drilling, linkage, highlighting, rotation, scrolling and other dynamic effects.

Not to mention provides a visual development platform, drag-and-drop component builds, basically zero code. Personalized chart needs, such as industrial 3D, park 3D, GIS and other projects, provide a complete interface to directly connect professional charts or plug-ins.

Compared to code development, data transfer, building the canvas, designing the layout, scaling components, component dynamics, etc., all require code development line by line and tuning back to the hardware.

In addition, the large screen needs to be put on the LED electronic screen, and the resolution of different screens is often different, so it is necessary to adjust the resolution during development to ensure the presentation effect, which is the most time-consuming, and it takes 50% of the debugging time before using code customization.

FineReport has an adaptive layout for large screens, which reduces development effort. This mechanism, once designed template, can support PC, tablet, mobile phone, large screen and other multi-terminal adaptive display.

Cause two: Data and platform performance are guaranteed on the back-end

Where to get the data from the big screen, how to maintain its quasi-accuracy, and how to comb the data around the core indicators of the business, these generally need to be backed by a data warehouse at the back end.

For example, I met a project that needed to take numbers from multiple data sources at the same time and do correlation calculation. They had designed the large screen page, but the data could not be retrieved because there was no back-end guarantee.

Data is not integrated, data is not cleaned by ETL, and some data sources even have no direct interface for file import.

In addition, the real-time display of data on the large screen has a high refresh frequency, which requires the data processing capacity of the background to be strong enough to ensure that the frequency of front-end data changes can meet the needs.

Sometimes the large screen also needs to display real-time data, so the historical data (usually in data warehouse) and real-time data (production system) need to be mixed and output to the large screen for efficient presentation.

Therefore, as a data front-end application, large screen should be able to connect with data warehouse or big data platform on the one hand, and also have high performance of computing response data on the other hand.

It takes 10 million levels of data to connect to a custom-developed large screen, and it takes 10 minutes to refresh a component. How much efficiency will be affected? If you want to develop adaptation performance in the backend, this amount of work and cost is not as good as using reporting tools directly.

Therefore, FineReport itself is a tool that supports docking with various data platforms. It not only has the ability of front-end report visualization, but also has the ability of back-end data computing engine, which can connect with high performance and big data. Why not?

Reason three: The development efficiency and maintenance cost are relatively low

As mentioned above, the high efficiency of developing large data screen with report tool FineReport is mainly reflected in:

Operation drag zero code:

  • Canvas interface, component drag-and-drop operation, component refresh (partial refresh), support component hiding, support component stacking, and independent expansion of different components.

  • A variety of layout methods: adaptive layout, absolute layout, Tab layout, support for multi-tab rotation,

  • Provide a variety of charts, controls, tables and other components, different components can correspond to different topics, bind different data sources, to achieve multidimensional analytical management cockpit

The last

In general, large screen to do well, the front-end effect mainly depends on custom development, back-end performance to choose suitable tools. As large screens become more routine, development efficiency and data performance assurance are more important factors than visualization.

If your company has the basis of data warehouse or data platform, while improving the report system, you might as well use reports to develop data large screen.