The technology for high–performance long–distance transmission has significant strategic value in the context of China's "East Data, West Calculation" project, which aims to construct a nationwide integrated computing network. Three trends are driving new requirements for the paradigm of wide–area distributed computing power coordination: 1) the rise of AI large–scale model intelligent applications that demand extremely high computing resources, 2) the embargo on high–performance GPU chips limiting single–center computing power resources, and 3) the formation of a dispersed computing power distribution pattern due to the establishment of computing clusters across various regions in China. High–performance long–distance transmission technology is crucial for this new paradigm. This paper discusses five aspects: supporting the new paradigm of wide–area distributed computing resource coordination, technical routes, underlay networks, challenges, and costs. Based on the results of a real–network experiment spanning 2100 kilometers from Shenzhen to Zhongwei in Ningxia, the authors believe that optimizing existing RDMA technology for long–distance transmission over wide–area optical networks is one of the most feasible and cost–effective solutions in the short term. By optimizing RoCE (RDMA over Converged Ethernet) over wide–area optical networks, it is possible to achieve "optical direct access to data in WAN" approaching physical layer communication performance indicators.

Ultra–high–definition (UHD) video is one of the key development directions for China's audiovisual industry. In response, relevant national departments have introduced a series of policies to encourage and support the development of the UHD video industry. The processes of capturing, transmitting, producing, and broadcasting UHD video, especially when integrated with advanced AI content generation technologies such as ChatGPT and Sora, exhibit typical characteristics of high bandwidth, high computing power, and low latency. These characteristics pose significant challenges to computing power and network infrastructure. Based on the typical requirements of UHD video and the latest trends in computing and network technologies, this paper proposes a computing power network architecture tailored for UHD video. It comprehensively employs key computing power network technologies such as heterogeneous computing resource networking and secure transmission technologies, UHD video service demand modeling and resource orchestration technologies, joint scheduling and routing planning technologies combining "data, computing, and modeling," and high–speed transmission technologies for UHD video. This approach aims to achieve the aggregation and networking of heterogeneous computing power across the country, thereby meeting the diverse computing and network transmission needs of various stages of UHD video processing, including capture, editing, and broadcasting.

As a security protection mechanism for convergence of computing and networking, security of networking and computing is a new technology and new engineering. Security of networking and computing involves zero trust, security access service edges, and blockchain of networking and computing and other technologies. It can provide a multi–dimensional security guarantee for the integration of the network, platform, application, and data in a security system. This article outlines the overall development trend of security of networking and computing, extracts the technical reference architecture of security of networking and computing, summarizes three typical application practices including security of computing and network convergence, security of cloud and network, and security of industry integration. Finally, combined with the severe challenges faced by the current security of networking and computing field, three aspects of development suggestions from standard system construction, application practice planning, and industrial ecological construction are given three aspects of development.

Computing-aware traffic routing(CATR), building upon traditional IP networks, collaboratively perceives computing resource information to provide efficient and flexible network capabilities, enabling the effective coordination of computing resource and networks. With technological advancements, it is gradually moving toward industry practice. This article elaborates on and analyzes the progress in technical research and key enabling technologies of CATR, such as independent semantic computing service identification and hierarchical CATR. Based on this, it proposes technical expansion directions for CATR, such as hybrid computing-network service function chains, and highlights the advantages of deploying CATR in business applications. It is recommended to leverage the outcomes of standardization efforts to further promote the implementation of CATR related technology in products and business applications.

Amidst the rapid expansion of large-scale models and the AI sector, epitomized by advancements like ChatGPT, there is a burgeoning demand for intelligent computing power to facilitate extensive distributed computing applications. Traditional data centers are facing challenges to accommodate the performance requisites of these scenarios. Currently, the Chinese government has already issued numerous policies to expedite the development of artificial intelligence data centers, purveying clear policy guidance and accelerated planning and construction blueprints. A high-performance network is pivotal within AIDC, serving as the backbone for computational tasks and enabling inter-data center connectivity and efficient data transmission.This paper aims to establish a robust technical framework to propel the continuous development of high-performance network by primarily investigating key technologies for high-performance networks in artificial intelligence data center (AIDC). Core requirements in transport protocols, networking, and operation administration and maintenance (OAM) for large-scale AI tasks are studied. Based on these demands, this paper further investigate the evolving demands on different layers of Artificial Intelligence Network and delves into core technologies, e.g. network architecture, congestion control policy, load balance policy, operation administration and maintenance. Subsequently, from the two major perspectives of network protocol development and all-optical networks, this paper analyzes the future developmental trends of AIDC networks. To establish a robust high-performance network framework within AIDC, this paper concludes that sufficient network performance, such as a near-lossless network environment, adequate interconnectivity, and solutions to storage performance bottlenecks in distributed storage scenarios, must be purveyed effectively. Furthermore, the development of high-performance networks in AIDC necessitates the integration and synergy of key technologies, including standardized networking schemes, innovative load balancing and congestion control protocols, and advanced OAM mechanisms, to enhance operational efficiency. High-performance AIDC networks must also offer comprehensive and universal device and resource awareness, allocation, scheduling, and OAM across the entire network, providing highperformance lossless transmission capabilities.