Since Ali first introduced the concept of "Middle Taiwan" in 2015, the Chinese market has rapidly become popular in China and data centers, so that Gartner added "Data Center" and "Data Center" in the "2020 China ICT Technology Maturity Curve" for the first time. "China-Taiwan Architecture" two emerging technologies. Gartner pointed out that taking into account the exponential growth of its users in the past few years, China Taiwan has become a key topic in the Chinese market. Many companies believe that China Taiwan is one of the key factors in accelerating digital transformation, but how to design a landable China Taiwan architecture But it is a challenge for many IT leaders in the market.
Data middle station is the hottest middle station technology among all kinds of middle station concepts. Gartner's "2020 China ICT Technology Maturity Curve" pointed out that in view of the overheated hype in the market and the inherent confusion about the definition and delivery of data in the platform, data in the platform technology is closer to the expected expansion period. Alibaba Group put forward the concept of data center, and many Internet companies have subsequently given their own definitions and products. If companies do not know how to build reusable data and analysis capabilities in their business environment, it will be difficult to be effective Use these capabilities.
The "expansion period of expectation" is actually very consistent with the current state of the Taiwanese market in the data. Many Taiwanese manufacturers in the data have put forward their own views, but they have created more data chimneys and data islands. It has just been selected as a typical application solution provider of IDC Enterprise Data Intelligence Implementation and Deployment Guide, and also selected by Gartner's "2020 China ICT Technology Maturity Curve" report. The data center methodology of the company solves the connection and conversion from data to intelligence, from traditional IT to new digital technology, from traditional business processes to new Internet services in the actual digital transformation process, without changing the existing IT of government and enterprise Under the premise, complete digital transformation.
On December 4, 2020, at the media exchange meeting of Minglue Technology "Let the Data Center Enter the Era of Digital Intelligence", Liu Guodong, the head of the data center of Minglue Technology, introduced this unique data centering methodology.
Three dimensions and two elements in data
"Digital transformation is not only the engine for companies to burst into new vitality in the new wave of technology, but also the company's strategic layout for the future." This was at the 2019 Data Intelligence Summit held by Minglue Technology Group in November 2019. Judgment proposed by Wu Minghui, founder, chairman and CEO of the Technology Group. At the summit at that time, Minglue Technology released the "New Generation Data Center" product, emphasizing the use of knowledge graphs as the carrier of the industry's Know-How, with the distinctive features of "data collaboration, co-construction and sharing, and reaching consensus". "Building corporate brains for industry customers" is the ultimate goal.
One year later, Liu Guodong, the person in charge of data center of Minglue Technology, summarized the practice of data center of Minglue Technology as the core technology system of data center, which has three dimensions: the first dimension is EAI enterprise application integration/enterprise process integration (IBPMS/RPA), the second dimension is heterogeneous computing/federal machine learning (HATP/FL), and the third dimension is object-oriented architecture (OOA). Liu Guodong emphasized that the data center technology system is a classic information engineering management system, which has existed for a long time, but its manifestation is different at each stage.
First of all, the core of the first dimension is integration. It draws on the classic enterprise-level architecture integration and business process integration in history. The integration technology capability is a large motherboard that supports the core of the data center. On this large motherboard, Plug in, Plug out and orchestrate various management capabilities, production capabilities, and collaboration capabilities; secondly, the second dimension is heterogeneous computing platform/federal machine learning, and the data center ultimately needs to complete or support heterogeneous computing systems and heterogeneous data mining And various effective coordination linkages of the analysis system; finally, the third dimension is the object-oriented architecture, which means two dimensions for the data center. One is that the users of the data center want to get a WYSIWYG The second is that the internal engineering system of the data center can ensure the high availability and scalability of technical modules through object-oriented design.
In addition to the three-dimensional degree of data center technology, Minglue Technology believes that the two core elements of data center are the full life cycle of data and enterprise-level federal middleware. Data life cycle management refers to four stages including access, integration, processing, and consumption. Among them, the core function of access is to access multiple heterogeneous data and distribute effective data assets; integration is to effectively sort and manage data assets; processing is based on algorithm models to generate data with new commercial value ; Consumption refers to the effective use of existing data assets.
The "federation" of enterprise-level federation middleware has two dimensions: the first dimension is the smooth connection between the new IT and DT capabilities of the enterprise and the traditional IT and DT capabilities, and finally achieves the goal of effective synergy; the second dimension is the right Heterogeneous computing capabilities, management capabilities, and organizational capabilities are coordinated orchestrated as needed. Liu Guodong emphasized that middleware is a link between the previous and the next, and the final effect is to ensure the relative stability of the back-end and the rapid iteration of the front-end data business.
The three-dimensionality of the data center technology system and the two major elements are Minglue Technology’s own unique data center methodology. Minglue Technology also provides complete data center products and technologies to help companies drive digital transformation through the construction of data centers . Liu Guodong emphasized that the data center is the backbone of the organization's digital transformation.
Leading digital transformation with data center
How to understand the methodology of data centering proposed by Minglue Technology? How to use data to organize and drive the digital transformation of government and enterprises?
Let's look at an actual case in detail.
As Tencent Cloud's partner in embracing the industrial Internet, Minglue Technology relies on the "two horizontal, multi-vertical and multi-center" strategy to carry out in-depth cooperation with Tencent Cloud in multiple fields to help government and enterprises respond to new infrastructure and accelerate the development of the digital economy. Relying on the powerful data management and knowledge graph capabilities of "Data Center" and "AI Center" (ie Liangheng), Minglue Technology provides basic components for Tencent Cloud in terms of underlying technology. For example, in the Changsha Urban Ultrain Data Center Project, 56 commissions, bureaus and bureaus have been connected in just three months, and 2.7 billion pieces of core platform data have been stored. Minglue Technology and Tencent Cloud together won a four hundred "battle" within three months: one hundred systems on the cloud, one hundred types of core data data inventory, one hundred government affairs integration services Landed, with one million active users, and through a series of solid data center support capabilities, helped the Changsha government lay a good foundation in the process of digital transformation.
So, how does the Changsha Super Brain Project reflect the three-dimensionality and two elements of the Taiwan-based technology system in Minglue's data? First, and perhaps the most puzzling is integration. Minglue Technology chose EAI enterprise application integration and enterprise process integration iBPMS and RPA. Anyone familiar with the history of enterprise IT technology knows that EAI and BPM are traditional enterprise business and process integration technologies, and iBPMS is enriched by adding artificial intelligence technology to BPMS. On the whole, EAI and iBPMS belong to the category of traditional business process integration, while RPA business process robots are the hottest business process integration technologies in recent years. What is the relationship between the three business process integration technologies of EAI, iBPMS, and RPA and the data center?
Liu Guodong explained in detail that many government and enterprise customers have old IT systems, as well as new DT systems based on Hadoop, and these systems are involved in core business systems. Therefore, when Minglue Technology is planning the data center, it tries to link the old and new business systems in terms of business process management, quickly orchestrate, and quickly collaborate to form new business processes, such as super brain business. Simply understand, the data center of Minglue Technology is not completely a data center in the literal sense, but provides data service middleware that supports smooth business transition, that is, without changing the original business system, The different processes of the original business are reconnected to form a new business process such as the super brain.
The second is heterogeneous computing/federal machine learning. For traditional government and enterprises, there are already different IT systems that use different computing engines, such as Oracle, MySQL, Hadoop and other new and old computing engines are being used at the same time; in addition, for example, the Changsha Ultrain project also needs to join Tencent Cloud TBDS And other new computing engines. When these different computing engines coexist in the IT environment of government and enterprise, it is a completely heterogeneous computing system. When computing tasks are distributed after the business process is choreographed, some of them must be pushed back to the original computing engine to complete the corresponding computing tasks, and some of them will run on a new computing engine like TBDS, and then use a concept similar to a bus. The calculation results are coordinated and linked.
The third is object-oriented design. Through object-oriented design, it provides good visualization capabilities for the use of new business systems and data centers. For example, based on the enhanced metadata system, the business and technical personnel of various commissions and bureaus in Changsha have intuitive insights into the entire system. Ability and can be directly related. The two-way mapping between the familiar business metadata system such as business terms and processes and the technical metadata system similar to the library table ensures that the business end users and the technical end users are connected at any time, and they have the ability to have comprehensive and in-depth business data insight. . For the newly added business process, there is no need to find another IT company to implement it. You only need to draw the business flow on the data center tool of Minglue Technology and import it into the system, which is the new business process conceived. When the business process is completed, you can also use the data center tool of Minglue Technology to associate with the corresponding database and table in a visual drag-and-drop manner for configuration.
Liu Guodong emphasized that the digital transformation of Changsha Ultrain with the data center is indeed a very complicated project, but it is always the same-the project is not subversion but compatibility and collaboration. Everyone often says that a technology stack is a set of technologies to form their own systems, which need to inherit traditional and classic technologies, but also need to collaborate with later new technologies. In the past, it was through the data lake method, but later proved that this road is suitable for downstream end analysis, but not suitable for organizational collaboration and consensus, so Minglue Technology adopts federal collaboration technology.
Data center to solve practical problems
It can be said that the data center of Minglue Technology is a set of methodology and a complete technical system summarized in actual combat. Using data center to drive digital transformation and support the connection and conversion of new and old businesses is the strategic layout of the enterprise facing the future.
The data center of Minglue Technology has solved many problems in the actual government and enterprise IT environment. For example, EAI, iBPMS, etc. are actually traditional enterprise technology integration interfaces. Why does Minglue Technology use these technologies? Liu Guodong explained in detail that the data business process management of Minglue Technology Data Center has the same form of expression as middleware in BPM, OA, and ERP in the past. The purpose of this is to make users feel the difference. However, the actual technology under the surface has actually changed significantly.
The data center of Minglue Technology is similar in interaction and form to the past interactive experience, which makes it easy for users to get started, because the interface is very familiar, but Minglue Technology can achieve multi-cluster, heterogeneous cluster, and heterogeneous cloud. The business process orchestration can quickly compile a new set of business processes on top of multiple clusters and heterogeneous clusters. For example, if the business of a certain city in the mainland is deployed on different clouds, Minglue Technology can realize heterogeneous collaboration and federal collaboration between different clouds, clouds and local IT, so that different technologies can be managed and work together.
The data center of Minglue Technology is used as the carrier of big data infrastructure. On the new big data engine, it chooses to cooperate with Tencent Cloud TBDS to realize the cloud-native data center system, which integrates Tencent's experience in massive data storage and computing And capabilities, it can provide 10,000-level cluster node scale, PB-level data storage and processing capabilities, second-level interactive query capabilities for hundreds of billions of data in thousands of dimensions, and platform high availability capabilities that have been verified by Tencent's 95% business. The advantages of both parties complement each other, and in terms of the overall capabilities of the data middle-office solution provided to users, it has solved the technical worries of users.
However, the reason why the data center of Minglue Technology chose federal collaboration instead of data lake, Liu Guodong emphasized that it is not that the data lake technology is not good. The data lake has its advantages, especially when the data is collected together, the engineering design and efficiency can be controlled. But the data lake often overlooks a problem-organizations and users live in reality, not in technology. As an organization, it is easy to compete for benefits. Therefore, data barriers cannot be solved, causing problems in the data lake. The reason for turning to federated coordination is that federation relies on the spirit of contract, which is a win-win situation.
Liu Guodong emphasized that the data center does not simply complete scheduling and process orchestration, but must link existing systems, existing systems, and new technical systems to interact with them. This is the challenge.
Overall, the bottom layer of the data center architecture of Minglue Technology is connected to multiple heterogeneous data sources, and the regular operation and management of data assets is completed through data standardization, data quality management, data security, enhanced metadata, and operation and maintenance management systems. . On top of this, through the enhancement of the data operating system, the value of Taiwan in the data is solved and developed. As a management system, the entire data life cycle manages the access, integration, processing, and consumption of data. The data asset operation platform above, comprehensively analyzes data assets from the operational dimension, including management systems such as data asset analysis, data asset maps, and data asset transactions.
In summary: In September 2020, the State-owned Assets Supervision and Administration Commission of the State Council formally issued the "Notice on Accelerating the Digital Transformation of State-owned Enterprises". As a key technology for the implementation of enterprise digitalization, the data center will help all industries to carry out the digital reconstruction of enterprises at full speed And transformation. As a technical system from actual combat, the data center of Minglue Technology is combined with the IT status quo of government and enterprises. On the basis of inheriting the existing system, the new system is associated and integrated, and the data center is truly implemented. Promote the digital transformation of government and enterprises through the construction of data centers, and help government and enterprises enter the era of digital intelligence down-to-earth. (Text/Ningchuan)