Data centers are important physical carriers that carry computing resources and provide computing services. They not only have the social service attributes of traditional infrastructure, but also have the empowerment attributes to promote the development of the digital economy. They are a new type of information infrastructure that countries around the world focus on developing. With the increasing enrichment of various artificial intelligence application scenarios and the rapid evolution of user needs, the development of data centers has ushered in new opportunities and challenges. This paper mainly studies the layout, types, scheduling, network, green, low-carbon, and industrial empowerment of data centers in order to provide reference for the development of computing technology and industry.
With the deepening of the project of national computing network to synergize east and west, how to improve the level of cross-regional computing power scheduling and provide high-quality computing services has become an important research direction of the computing industry. This paper focuses on the problem of computing power scheduling in the computing power network, focuses on the architecture and key technologies of computing power scheduling, and introduces the overall implementation path of computing power scheduling from three aspects: integrating computing resources, building a computing power scheduling platform, and establishing a standard and specification system.
The computing power network is a new type of intelligent network, and its infrastructure includes communication networks and heterogeneous computing power facilities. By integrating multiple resources, such as data, computing and network for orchestration and management, the integration of network, computing power, data, operation and maintenance, intelligence and services is realized. A high-quality computing power network is a powerful booster for the vigorous development of the digital economy. It will empower the digital and intelligent transformation of thousands of industries, and will also become an important direction for the integration and innovation of information and communication technologies. This paper introduces the current situation of computing power networks at home and abroad, points out the main challenges faced by the current development of computing power network, and finally puts forward a set of evaluation system based on the development stage of computing power networks.
To address the problem of measurement and modeling of computing power networks, this paper first reviews the current research progress from the narrow and broad definitions of measurement system for computing power network, including the measurement of computing network resources and the measurement of computing network demand. Then, this paper proposes a measurement and modeling scheme for service-centric computing power network from the perspective of computing power network services. Finally, it summarizes and discusses the problems and challenges of this scheme.
Cloud-network convergence is an inevitable trend of the deep convergence of communication technology and information technology. National policies, markets and technologies are driving the evolution of cloud-network architecture from cloud-network collaboration to cloud-network convergence architecture. The new data center represented by cloud-network convergence is the development direction of the industry. Along with the expansion and development of businesses, the innovation of key technologies is urgently needed. In this paper, from the evolution trend of data center for cloud-network convergence, new services and new challenges are elaborated, and key technologies of the new generation of data center for cloud-network convergence are studied and analyzed.
Agaist the backdrop of the new round of global information technology transformation trend, improving the circulation level of data factors has become the consensus around the world. The rational use of computing power networks to realize the on-demand allocation and flexible scheduling of computing and storage resources between clouds, edges, and devices is the guarantee for effectively releasing the value of data. This paper first comprehensively sorts out the policies related to the storage industry in some major countries and regions around the world. Then, it analyzes the current situation of the storage market, as well as the development environment, the storage capacity and innovation capabilities of China’s storage industry. Finally, it looks forward to the development trend of China’s data center storage capacity and puts forward suggestions for high-quality development.
Under the background of the project of national computing network to synergize east and west, the computing power infrastructure has ushered in rapid development, resulting in a large number of heterogeneous computing power resources. The computing power and network collaboration of cloud, edge, and device can meet the efficient matching of multiple heterogeneous computing power resources with many application scenarios. Based on the research of heterogeneous computing power network scheduling, this paper provides a research idea for the scheduling technology to achieve heterogeneous computing and network convergence, systematically analyzes the technological development status, key technologies and challenges, and points out many application scenarios. However, in practical applications, heterogeneous computing network scheduling is more complex, and it still faces many uncertainties and challenges, which requires continuous improvement of technology.
With the acceleration of digital transformation of business applications, the requirements for dedicated, flexible, ubiquitous and collaborative service capability of computing infrastructure for various digital scenarios are also increasing. The computing and network convergence is another upgrade of computing network supply mode after coordinated development of data center and network as well as cloud-network convergence. This paper focuses on the analysis of the business requirements, industry status and development trend of the computing and network convergence, and further proposes suggestions for the development of computing and network convergence against the backdrop of building a national computing network to synergize east and west.
The computing and network convergence has become an important measure for accelerating the digital transformation of industries because it can provide high-efficiency computing power, advanced storage capacity, and high-speed transportation capacity. The evolution of computing and network convergence is also accompanied by challenges such as multi-cloud collaboration, cloud-network collaboration, cloud-data collaboration, and storage-computing collaboration. At present, the demand for migrating to the cloud in all walks of life is soaring, but most of them choose the hybrid multi-cloud deployment method, and data storage is used as the business data carrier. How to coordinate the efficient development of data storage and multi-cloud to promote the implementation of computing and network convergence has become more and more important. This paper analyzes the development trend of data storage in the process of multi-cloud development, and the challenges faced in data management and data flow in a multi-cloud environment. It proposes some suggestions on the coordinated development of data storage and cloud native so as to help enterprises to choose the appropriate storage type under the multi-cloud architecture.
Based on the development of the data center industry, starting from the perspective of the whole process from the generation to the use of computing power, this paper first analyzes the development status and key directions of hot issues such as the use of renewable energy power, the implementation of diversified evaluation indicators and the improvement of the utilization efficiency of computing platform resources in the data center industry. Then, it summarizes the current situation and predicts the trend of the data center computing power network and power grid convergence from the perspective of green and low-carbon development. Finally, it proposes the development prospect.
Under the guidance of the carbon peaking and carbon neutrality goals, it is crucial to reduce the energy consumption of computing and promote “green computing”. Immersion cooling has been widely concerned in recent years due to its outstanding advantages of heat dissipation and energy saving. The premise for data centers to pursue high energy efficiency is to ensure the high reliability and availability of data centers. The security, stability, and sustainable evolution of data centers are more important indicators than energy saving. In order to meet large-scale applications, it is necessary to study and evaluate the key technologies such as reliability and evolution capability of the system architecture. From the perspective of the whole process, this paper sorts out the reliability levels and energy efficiency of different liquid cooling technologies by starting from the overall situation of the data center and IT equipment. Then, it focuses on the analysis on reliability and evolution capability of the single-phase immersion cooling technology. It is expected to provide a reference for the large-scale application and development of the single-phase immersion cooling industry in China’s data center.
This paper starts from clarifying the basic concepts and roles of data sharing, analyzes the current situation of data sharing management in major countries and regions, and explores the differences and connections between the target positioning and key factors of data sharing management in various countries and regions. Then, it analyzes the situation of data sharing management in China. Finally, based on the basic situation of data sharing, international management experience, and China’s current situation, this paper puts forward the overall idea and suggestions for data sharing management in the telecommunications and Internet industries.
In the context of building “new infrastructure” and smart cities during the 14th Five-Year Plan period, the current situation, problems and trend of the urban perception system are analyzed. Moreover, the technical framework of the perception system and the overall suggestions for further development are put forward based on relevant practices and cases in China.
Based on the layered architecture of the Internet of Things (IoT), this paper analyzes the security risks and security needs faced by each layer of the IoT in the era of intelligent interconnection. The layered security architecture and corresponding security measures of the IoT are proposed, forming the overall security idea of IoT, including basic security capability, sensing security, network security, platform security, application security and security protection. In addition, the organization, content and progress of the IoT security standards are studied from both international and domestic aspects, which gives a research direction for IoT security standards and further promotes the building and improvement of intelligent IoT security system.
At present, the smart elderly care system in China is faced with technical problems in realizing the accurate data collection and fast communication connection without latency, and overcoming insufficient artificial intelligence intervention. This paper designs a smart elderly care service platform based on emerging technologies such as 5G network, artificial intelligence, and edge computing. This platform provides low-latency and highly reliable communication connection for elderly care business through the customized private 5G network, and realizes the interconnection of data across systems and terminals, which promotes intelligent collaboration between elderly-care service resources and service systems.