Shulou(Shulou.com)06/02 Report--

In the era of industrial Internet, the digital transformation of enterprises will become a trend. A previous survey by IDC, a well-known global research institution, shows that by 2018, 67 per cent of the world's top 1000 companies and 50 per cent of China's top 1000 companies will regard digital transformation as the strategic core of their enterprises; by 2020, 20 per cent of China's GDP will come from the added value of business digitization transformation, which will rise to the macroeconomic level, reshaping the economy while changing the way companies operate.

In fact, digital transformation is to integrate the application of digital technology into the internal management field and the external changing business environment, thus having a decisive change to the whole business value chain. So how can digital technology help enterprises to carry out digital transformation? Then we should start with the middle station.

1. What is Zhongtai

Zhongtai is a new concept, but it is an old term, but we give it a new connotation in the new era. This is because in the era of digital economy, great changes have taken place in the way of IT construction. In order to respond to the demand quickly, we need to use innovative ideas to build IT.

Zhongtai, through the abstraction of business, data and technology, reuses the service capability, constructs the enterprise-level service capability, and removes the barriers among the internal business departments and branches at all levels of the enterprise. adapt to the development strategy of business diversification of enterprises, especially large enterprise groups. Based on Zhongtai, we can quickly build foreground applications for end customers, so as to meet the foreground needs of various personalized features, and provide a clear path for the digital transformation of enterprises.

Definition of Zhongtai:

Zhongtai is an enterprise-level capability reuse platform, which has a common capability and supports multiple services. The core is "functional reuse", to build "large and medium-sized platform, small foreground" to meet the needs of rapid business expansion.

two。 The main responsibility is to collect all business data, cooperate with various business units, refine the common needs of the business, and support the rapid development of front-end and back-end business.

3. It precipitates a large amount of user behavior data (including non-system users), which lays the foundation for the new business model of big data's intelligent algorithm.

4. As a provider of business services, it does not need to rely on the stability of the business, but needs to provide continuous capacity support for new services.

Second, the relationship between the middle stage, the foreground and the backstage

To understand the central station, we must first understand what the foreground and backstage mean respectively:

Foreground: a front-end platform composed of all kinds of foreground systems. The foreground is the front-end platform of the system and the application layer that directly interacts with the end users. Each foreground system is a user contact, that is, the end user of the enterprise directly uses or interacts with the system, which is the intersection between the enterprise and the end user. For example, our daily use of APP, H5, PC and Mini Program all belong to the foreground system.

Background: a back-end platform composed of background systems. The background refers to the back-end platform of the system, and the end user cannot perceive its existence. The value of the background is to store and calculate the core data of the enterprise (data + computing), such as financial system, product system, customer management system, logistics management system and so on.

Infrastructure and computing platform, as the core computing resources of enterprises, are also a part of the background. But the background is not born for the foreground, because the change of the front end needs the change of the back end to support, so the rapid response of the background is required. However, the goal of setting up the background is not to serve the foreground, the value of the background is to store and calculate the core data of the enterprise, and to improve the security and stability of the back-end data and the management efficiency of the system.

At this time, the foreground and the backstage are like two gears with different rotational speeds. In order to quickly respond to the needs of front-end users, the foreground pays attention to rapid innovation and iteration, so the faster the speed, the better. In the background, due to the relatively stable back-end resources, and often the system is outdated and complex, and even subject to laws and regulations audit and other relevant compliance constraints, so it is the supremacy of stability, the more stable the better, the slower the speed, the better. Therefore, with the continuous development of enterprise business, the problem of "matching imbalance" of the gear speed of "foreground + background" gradually appears.

In order to change this contradictory situation and adapt to the rapid iteration of demand, the "Zhongtai" was born, and the Zhongtai system is like a gear, driving the front and background systems to rotate rapidly. We can separate the logic layer of the background to form the product architecture of "foreground (application layer)-middle platform (logic layer)-background (data layer)". Under this product architecture, when the foreground demand comes, the middle station can respond quickly, thus improving the efficiency of research and development and reducing the cost of innovation.

III. Classification of China Central Station

The middle platform is generated from the requirement of sharing multiple similar foreground business applications, so the first proposed middle platform is the business center, and the data is generated from the business system, and the business system also needs the results of data analysis. can the data storage and computing power of the business system be separated and the storage and computing power provided by a separate data processing platform? This not only simplifies the complexity of the business system itself, but also allows each system to adopt more appropriate technologies and focus on what the province is good at. This dedicated data processing platform is the data center.

Of course, there is also the technology center, which provides a simple and consistent application interface, integrates and packages the capabilities of various technical middleware, and provides unified technical support and output to the business. Finally, there is the middle platform of the organization, but the adjustment of the organizational structure is a big problem. The typical organizational structure includes: straight line functional type (U), business department type (M), matrix type, network type, platform type, not counting various combinations and variants; if combined with the business model, we also need to understand the amoeba business model and Haier's independent business model. Therefore, these are certainly not clear in this article, so this article focuses on the business center and the data center.

IV. Business center and data center

The business center is the transformation of the enterprise IT business architecture first put forward by Alibaba. From the overall point of view of Alibaba, the business center is planned as a whole from the aspects of overall strategy, business support, connecting consumers and business innovation. Ali's business center pays more attention to how to support and serve the business.

Ali deploys the technology console through the Aliyun platform, provides support for shared business units within the group, and ultimately provides service-oriented capability output to the front desk business lines.

To put it bluntly, the business center is to separate the business from the business logic, and clearly describe what services, data and functions they have through the formulation of standards and specifications, so as to reduce communication costs and improve collaboration efficiency. let any business line have the core competence of the whole company, and provide business parties with the ability to innovate quickly and at a low cost.

From a technical point of view, the central platform is to build a flexible and rapid response to changes in the framework, faster to achieve the needs of the front end, to avoid a high degree of reuse of the repeated construction of functions, this is agile development, improve efficiency. From a business perspective, with the help of Zhongtai precipitation capability, we can support rapid innovation, make R & D more flexible and business more agile, in order to cope with unpredictable market changes in the future. To say the least, some functions and other business sectors have been done, so the bottom layer can be combined to make it more flexible and fast.

The business center platform is the core of the middle platform construction of the enterprise, and it needs to solve the complex coordination of multi-business areas, so it is faced with a great challenge, which requires the leader to take the lead, overall planning, step-by-step implementation, finding the right entry point, and defining the business goal and scope. First pilot, use the proven technology platform and mid-Taiwan methodology to enable business In the pilot process, continuous integration and iterative optimization, run-in their own concept of Taiwan, build their own business capacity ecology, can be gradually spread out, and finally optimize the organization and improve efficiency.

Next, what is the data center? What is the difference between data center and data warehouse? How to integrate the data center and the business center?

The so-called data center, that is, the realization of data layering and horizontal decoupling, precipitation of public data capacity, it can be divided into three layers, data model, data service and data development. Realize cross-domain data integration and knowledge precipitation through data modeling; encapsulate and open data through data services to meet the requirements of upper-level applications quickly and flexibly; and meet the needs of personalized data and applications through data development tools. See the following figure (take an operator as an example):

1. Data model

The data model is hierarchical and used to be called the data warehouse model. Here is summarized into three layers, the basic model is generally relational modeling, mainly to achieve data standardization, we call "book with text, car with track". The fusion model is generally dimensional modeling, which mainly realizes the integration of cross-data, and the form of integration can be summary, association, and parsing. The mining model is actually partial to application, but if more people use it, you can also precipitate the mining model to the middle platform as the knowledge of the enterprise. For example, the model of customer browsing preference analysis has great commonness, so someone should normalize it to the middle platform model. In order to open up for other people to use, CCTV is relative, there is no absolute standard.

2. Data service

The service encapsulation of the data model according to the application requirements constitutes the data service, which is exactly the same as the service concept in the business, except that the data encapsulation is a little more difficult than the general function encapsulation. After all, the change of the function of OLTP is limited, and the data analysis is greatly affected by market factors, which leads to the difficulty of service encapsulation. With the deepening of the operation of enterprise big data, all kinds of big data applications emerge one after another, and the demand for data services is very urgent. If big data does not serve, he will not be able to scale and meet the requirements of big data services inside and outside.

3. Data development

Only data models and data services are far from enough, because no matter how good off-the-shelf data and services are often unable to meet the front-end personalized requirements, it is better to teach people to fish than to teach people to fish. The last layer of the data center is data development, which is divided into three levels according to the difficulty of development. The simplest thing is to provide a tag library (DMP). Users can quickly form a marketing customer base based on tag assembly. Generally for business personnel The second is to provide a data development platform, based on which users can access all the data and carry out visual development, generally for SQL developers, and finally provide application environment and components, so that technicians can create personalized data products on their own. The above layers are progressive to meet the requirements of people at different levels.

5. Thinking behind the China Central Station

The construction of China Taiwan in enterprises is a systematic project, which can not be achieved in one step and can not be achieved overnight. We have seen Alibaba's construction of China Taiwan is very successful, and we have also heard different success cases, but the success of others does not mean our own success. The landing of the enterprise in Taiwan, behind the governance, is the enterprise architecture of keeping pace with the times and continuous iteration.

1. Strategic alignment

Ensure the alignment of business strategy and IT strategy, IT for corporate strategy, business output structural support and technical support.

2. Enterprise architecture

Enterprise architecture needs to have an insight into the changes of environment and industry, foresee the development of enterprise operation, establish a scientific and reasonable business architecture and IT architecture, and govern the relationship between IT assets and business and the organization, so as to ensure the IT alignment strategy and become the basis for the implementation of the strategy.

3. Reusable

If there are basic services, do you provide "corresponding services" or a "package of services" to the front desk through the central desk? It depends on the extent to which the service is open and shared. As we can see, the major Internet decides the beginning of the construction of China Taiwan, which is always accompanied by the adjustment of the organizational structure at the enterprise level. After all, the principle of Zhongtai construction is to break down the department wall, business wall and data wall and achieve the goal of "reusability" through open sharing.

4. Flexible expansion

After a package of mid-Taiwan services can be exported, the volume of business may soar in a short period of time. Whether it can withstand the high concurrency and high availability during the peak period of large traffic will become a big challenge. The flexibility of the underlying scalability will be very important. Enterprises should apply advanced development technology concepts such as DevOps, Docker and micro-services to start digital technology transformation before the construction of China Central Taiwan.

VI. The Future Prospect of Enterprise China and Taiwan

The growth of CCTV is inseparable from the innovation of the front desk business. Only by constantly iterating the business and updating trial and error, and putting forward new challenges and higher requirements to CCTV, can we make CCTV do better. On the other hand, China and Taiwan also need to have their own products, platform construction thinking, and as an incubator of new business.

China and Taiwan also need to build an open system. Opening up is not only open to the internal enterprise, but also to the outside of the enterprise. Through the construction of Zhongtai, enterprises can turn their own Zhongtai system into an open platform, so as to fully provide third-party data and services for other enterprises. In addition, China Central Station itself can also use and integrate third-party data and services through openness, and support more personalized needs and customized expansion through open API.

"there is no best, only the most suitable"! The most suitable enterprise platform must be when you have an in-depth understanding of the business, products, systems, and organizations, and not only understand where it is today, but also understand how it evolved in the past and how it will evolve in the future. Only when you understand all these things, can you make the best enterprise architecture design.

My blog will soon be synchronized to Tencent Cloud + Community, and you are invited to join us: https://cloud.tencent.com/developer/support-plan?invite_code=3hcvknlp9zqc0

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.