-
2022 Year 4 Month, Chengdu's first edge computing center—the Yunyo Technology Kunpeng Ecological Edge Computing (Chengdu) Operation Center—was put into operation;
-
2022 Year 8 Month, Zhangjiajie Sixi branch company located in 110 The distributed small data center cabin in the 1000 kV Wulingyuan substation was put into commercial operation ；
-
2022 Year 11 Month 10 Day, innovation demonstration “985 Project ” The distributed data center business model Hangzhou's first pilot project was officially unveiled at the Hangzhou Qianlong Electric Co., Ltd. substation ...
In the past year alone, the implementation of edge data center applications in China has sprung up like mushrooms after a spring rain. On the other hand, under the background of "East Data West Calculation", more and more large and super-large data centers are being built in 8 large nodes, 10 Large clusters are "rising from the ground". An invisible computing power network has begun to be woven. Data centers are gradually developing towards bipolarization—on the edge, small and micro data centers; within the cluster, large and super-large data centers.
The "conflict" between "physical" and "logical"
On the one hand, data center construction shows a "bipolar" trend; on the other hand, users want a logically unified virtual resource pool to achieve faster and more convenient business deployment. The "conflict" between "physical" and "logical" is already apparent. How to better achieve interconnection between data centers, and achieve flexible and rapid scheduling and interconnection of data and computing power among multiple data centers? Distributed data centers may become the "optimal solution" to the above problems. Distributed data center is actually more suitable to be called distributed cloud data center. Its essence is based on cloud computing technology, using thousands of servers, through virtualization technology, to form a huge virtual host, to achieve unified scheduling and management of data and computing power, avoid repeated investment, and better utilize hardware, software, and network resources. "Physically decentralized construction, logically unified management" —This is the simplest description of a distributed data center. At the "physical" level, data centers can be built anywhere, even underwater or in space; at the "logical" level, virtualization technology is used to allow data centers built in various locations to work better together, allowing users to use data centers to conduct business as easily as using a home computer. From the current development of the data center industry, it is not difficult to see that users' demand for computing power is increasing. The most ideal state is to build a large enough data center to meet users' demand for huge computing power, but this is only an "ideal". In reality, users need to build multiple data centers in the same place, or even in different places, to meet their needs for computing power and different data (cold data, hot data) computation and storage. At this time, Distributed architecture becomes the key to solving the "conflict" between "physical" and "logical". Distributed data centers can be divided into two categories from the application level— Single data center distributed architecture, multi-data center distributed architecture.
Single data center distributed architecture
This type of architecture mainly integrates distributed computing, storage, and network technologies, mainly used within a single data center, using distributed and virtualization technologies to maximize server utilization and improve efficiency.
Multi-data center distributed architecture
This type of architecture is mainly used in multiple data centers located in different geographical locations to achieve unified management of multiple data centers. Its core technology is still cloud computing technology, and wide area network interconnection technology. Multiple data centers are built into a distributed "virtual resource pool" that integrates computing, storage, and networking. This resource pool can unify and manage multiple data centers in different regions, and even serve as disaster recovery, with the characteristics of improving enterprise efficiency, reducing enterprise data center rental costs, and reducing security risks.
Whether it is a single data center distributed architecture or a multi-data center distributed architecture, in my opinion, when data centers enter a critical period of digital transformation, distributed architecture becomes the core of whether edge data centers and the "East Data West Calculation" project can "become a reality".
Why distributed architecture?
Why is distributed architecture an indispensable support in the "bipolar" development of data centers? Because based on cloud computing related technologies, data centers using distributed architecture can solve the "conflict" between the "physical" and "logical" levels of traditional data centers, allowing multiple data centers in different locations to form a large "computer". In addition to this advantage, distributed architecture has three major advantages compared to traditional data center architecture.
With the advancement of enterprise digitalization, its business is increasingly dependent on data centers. Once the data center "crashes", the loss to the enterprise is immeasurable. If a distributed architecture is adopted, enterprises can easily synchronize data backups to multiple, geographically dispersed data centers, on the one hand avoiding the risk of power outages or crashes and ensuring business continuity; On the other hand, due to their different geographical locations, it can also avoid the impact of "natural and man-made disasters" such as earthquakes, floods, and operational errors on enterprise business. For example: 2022 Year 10 Month 15 Day, due to SK C&C Banqiao data center fire, resulting in Kakao 's chat software Kakao talk and other applications were interrupted, affecting a large number of South Korean users ；2022 Year 12 Month 18 Day, Alibaba Cloud's Hong Kong data center node malfunctioned, resulting in the websites of key infrastructure operators such as the Macao Monetary Authority, Galaxy Entertainment, Lotus Television, and Macao Cement Factory, as well as Omik and MFood and other takeaway platforms, and Macao Daily, etc. APP Once inaccessible ……… If a multi-data center distributed architecture is adopted, the impact on users after a disaster can be minimized. Currently, the financial industry is the most mature in the application of distributed data centers. As we all know, due to regulatory requirements and its own requirements for data security and business continuity, the multi-center distributed architecture scenario is " City disaster recovery center " to " Two-place three-center active-active " architecture ( That is, the production data center, the city disaster recovery center, and the remote disaster recovery center ) evolution. With the development of the business, it may evolve into " Multi-place multi-center distributed multi-active " architecture. At present, the vast majority of banks regard data as assets. Through distributed technology, realizing data asset management and stimulating data potential has become an important means of improving the overall competitiveness of banks.
Distributed data centers allow users to choose service providers, not limited to one, and users can focus more on their own business development without having to worry too much about the details of underlying technology operation and maintenance and data center management. On the one hand, it helps users save the cost of operation and maintenance personnel. On the other hand, on-demand deployment and on-demand procurement, for critical core data and non-critical data can be purchased and managed separately, which can also make every penny of the user's money spent on the "blade", achieving real cost reduction and efficiency increase. Theoretically, users are not limited to using the services of a single data center operator. For example, you can use a virtual server in a Swiss data center, a server in a British data center, and a server in a Russian data center. Furthermore, under the background of "East Data West Calculation", at present, the "computing power cost" in the western region is obviously lower than that in the eastern region. Some non-real-time data processing, such as video rendering, machine learning, and large-scale data storage, enterprises can choose to process them in the western region. Through distributed technology, it can achieve smoother scheduling of these data and computing power, and further reduce the cost of enterprise leasing services.
As we all know, in the Internet age, the demand for speed is extremely high, and sometimes business data streams need to be calculated in milliseconds or even smaller time units. For example, in stock trading. Even millisecond-level delays can give competitors a significant competitive advantage, which competitors can use to buy and sell stocks earlier. For this reason, securities traders choose to use data centers in Frankfurt and London near major exchanges. For example, with the gradual popularization and application of autonomous driving technology, it is necessary to perform rapid calculations on data centers at the edge to ensure that autonomous driving vehicles can respond to emergencies and complex road conditions in real time, ensuring the safety of passengers and surrounding people and objects. At this time, Using distributed technology, making the data center closer to the user side and the edge side, can better schedule data and computing power, The circulation and calculation between various "nodes" are faster and more efficient compared with traditional centralized management. Four major challenges, cannot be ignored
Although distributed data centers have advantages such as better security and helping enterprises reduce costs and increase efficiency, in terms of current technology, there are four major challenges that distributed data centers need to face in order to be maturely applied.
I. Computing power scheduling
The management of computing resources is a current problem. How to efficiently schedule computing resources in response to sudden surges in business data flow in data centers located in different locations, and realize unified management of data in different geographical locations will be a major challenge faced by distributed data centers in terms of computing power scheduling. Using a distributed data sharing governance model to "integrate" centralized and distributed data governance and achieve coexistence is one of the more common and mature methods currently.
II. Cloud-network integration The core foundation of realizing a multi-data center distributed architecture is the network, and doing a good job of cloud-network integration is also the key to the success of "East Data West Calculation". As mentioned earlier, realizing "logically unified management" determines whether distributed data centers can be applied and implemented. Based in China, the main force in promoting the implementation of cloud-network integration is the three major operators. At present, and for a long time to come, the three major operators will still be the main force in China's network construction.
How to achieve data collaboration is one of the main challenges currently faced. With As business coverage increases, the importance of data center collaboration in various places is increasing. However, due to difficulties in distance and scale, the network bandwidth between data centers in various places cannot guarantee real-time data synchronization, which has a certain impact on data consistency and integrity and business continuity. In mid-2022, Gartner released the 2022 Storage and Data Protection Technology Maturity Curve, and software-defined storage infrastructure became the only transformative technology this year. Distributed file systems, object storage, and container-native storage all have a high priority rating. It is obvious that Distributed storage will become the next market wave for storage vendors to compete for 。
Security has always been one of the most important issues for enterprises. Although distributed data centers are more flexible than traditional data centers, how to ensure data asset security from both physical and virtual levels has become a major challenge. Under the distributed architecture, data will flow between multiple data centers. As the distributed data center architecture becomes more complex, the challenges to security are also increasing. In the process of circulation, how to build firewalls, build security protocols, and ensure data security has become a major challenge faced in the application process of distributed architecture. It is reported that a complete and unified security product solution has not yet been formed.
With the development of the data center industry in recent years, the data center has gradually developed towards "polarization", which is an undeniable fact. Distributed technology will play a crucial role in both the construction and application of small data centers at the edge and in the overall development of large and super-large data centers within the cluster.
At this stage, its technological advantages and challenges are obvious. With the development of 5G, the Internet of Things, and industrial internet technologies, I believe that 2023 will be a year of breakthroughs for distributed data center technology.
| Article Source: China IDC Circle
• END •
Disclaimer: Some of the publicly available information collected on this website comes from the internet. The purpose of reprinting is to convey more information and for online sharing. It does not represent this site's agreement with its views or responsibility for its authenticity, nor does it constitute any other suggestion. The content of the article is for reference only. If you find any works on the website that infringe your intellectual property rights, please contact us, and we will modify or delete them in a timely manner.
