It’s a new digital age in information service, traditional architecture detaching cloud and network cannot satisfied the demand of the integrated cloud-network service coming with the booming of the digital economy. As a foundation of new information infrastructure, cloud-network convergence is the inevitable choice of global digitalization and underlying trend in the development of the CT and IT technology. In this paper, we analyzes the background and development trends of cloud-network convergence and the development logic deeply, proposes three characteristics of cloud-network convergence. This paper highlights the cutting-edge technologies in cloud-network convergence research, and explores the development vision of new service, new network, new computing and new control and manaegment. Finally we presents the challenges of cloud-network convergence.
As a solid foundation in the era of digital economy, the digital information infrastructure has the cloud-network convergence feature of network as the foundation, cloud as the core, network moving with the cloud, and cloud network integration. Based on the evolution of existing network technologies and the convergence innovation trend of infrastructure, this paper proposes a new super convergence networking scheme with the characteristics of fixed-mobile convergence, computing-network convergence and cloud-optical convergence. The proposed scheme with the cloud as the link realizes the connection of the access network, cloud access network, cloud intranet and inter cloud network, which provides an idea for promoting the integrated development of cloud and network. Meanwhile, each part of the scheme can develop independently, flexibly adapt to business requirements, and naturally adapt to the complex and uncertain evolution of new business in the future.
As a new paradigm of cloud-network integration, computing power network has received extensive attention from academia and industry. With the development of cutting-edge network technologies such as deterministic network, cloud native, artificial intelligence, etc., the development of computing power network also presents new trend characteristics. And it is urgent to design a new computing power network architecture to meet the needs of new business development. From the three dimensions of resources, time delay and intelligence, this paper expounds the development trend of computing-network cloud native, computing-network determinacy, computing-network self-intelligence, designs a new computing power network architecture, and analyzes and discusses related technologies and challenges.
At present, a new round of scientific and technological revolution and industrial transformation are in-depth development. The new infrastructure represented by 5G, Gigabit pipe network and Internet of Things is deeply integrated with the economy and society to become an important engine for promoting the digital, networked and intelligent transformation of the economy and society. This paper introduces the computing network integration in frastructure with the deep integration of “computing” + “network”, puts forward the reference framework of convergence of compute and network infrastructure, analyzes the capability of convergence of compute and network infrastructure from the aspects of key technology, industrial application and industrial ecology, summarizes the current industrial ecological structure, puts forward the development opportunities and challenges of convergence of compute and network infrastructure, and summarizes the ecological construction tasks.
In order to support important applications of smart cities and smart industry, future 6G mobile communication networks have to accommodate a tremendous volume of data traffic from sensing devices and base stations. Moreover, they also have to provide the computational capability for data cleansing and data processing for the sake of achieving accurate environment and state sensing. However, data collection and environment sensing consume huge communication and computation energy both on the network side and on the device side. It severely damages the sustainability of communication, computation and sensing tasks. In this article, we aim for reducing the energy consumption of communication, computation and sensing and for realizing green 6G networks. To this end, we briefly introduce some key enabling techniques, such as shortening the computation process, flexible networking and active wireless power transfer. We hope that it may intrigue more research interests in this fascinating topic.
With the development of cloud network integration architecture, cloud network slicing realizes the unified management and allocation of network resources, but the arrangement of cloud network slicing sharing resources exposes the risk of information leakage. This paper proposes a slicing algorithm based on resource sharing and user security rating, which maximizes the benefits of the slicing and makes the slicing meet the requirements of a certain mapping security level, and ensures that the deployed slicing meets the requirements of security performance. Experiments show that this algorithm can improve the satisfaction rate, resource utilization rate and cost profit rate of the orchestration.
As industry digitalization continues to deepen, enterprise services are evolving from cloud-based internet applications to cloud-based information systems, and production systems are gradually deployed on the cloud. After enterprise core production systems (such as smart manufacturing, remote control, and telemedicine) are migrated to the cloud, network indicators such as latency and jitter have clear requirements. Traditional best-effort methods cannot meet these requirements. The intelligent cloud network solution provided by carriers for enterprises should not only improve connection capabilities to adapt to service scope expansion, but also provide deterministic experience capabilities to meet production service requirements. This paper discusses the service requirements of intelligent cloud network, analyzes the technologies related to network deterministic experience, and defines the possible evolution stages of deterministic network, and gives the related experimental data.
In the era of computing power, heterogeneous data processing of new computing power services requires the ubiquitous computing power supported by cloud edge collaboration. Meanwhile, computing power is evolving to the new computing power including computing power, storage power, transportation power and capability, so as to meet the diversified demands of new computing power services. This requires the cloud-network convergence orchestrator and scheduler system to be equipped with automatic, intelligent and digital twinization capabilities of new computing power, so as to figure out the quantitative relationship between target SLA (Service Level Agreement) and resource portfolio of cloud and network conditionally dependent on the global topology of cloud-network convergence, in order to improve computing efficiency. On the basis of analyzing the deficiencies of the existing cloud-network convergence orchestrator and scheduler system, proposes a new cloud network convergence orchestrator and scheduler system architecture which is automatic, intelligent and digital twinization, gives typical scenarios, and puts forward suggestions on the evolution path of telco’s cloud-network convergence orchestrator and scheduler system.
Cloud-network convergence relies on the aggregation capabilities of the cloud to realize the collaboration, integration and innovation of multiple technical elements, and has become the best choice for the scenario-based needs of vertical industries. Massive data growth, frequent data interaction, and the slowdown of Moore’s Law all place higher requirements on the performance and cost of infrastructure, and smart NICs emerge as the times require. This paper compares the two acceleration processing modes centered on computing and data, analyzes the academic research progress of smart NICs, the characteristics and applications of mainstream products, and looks forward to the technological development trend of smart NICs. The analysis results show that the intelligent network card can maximize the overall performance of the system and “reduce the burden” for the hardware processor.
This paper analyzes the specific needs of the construction of computing network in terms of counting from the east to the west, through an overview of the development of the counting from the east to the west project, and expounds the solution of CUBE-Net 3.0 computing power network architecture. It also gives suggestions on the construction and development of computing power network under the background of counting from the east to the west in China, aiming to realize the research and analysis of the development of computing power network under the background of counting from the east to the west project construction.
This paper introduces the background of the strategy of east-data-west-computing project in China, and analyzes the evolution steps of cloud-netwonk convergence, as well as the multi-cloud management technology and functional characteristics. Combined with the landing scene, this paper discusses the role of multi-cloud management technology, and finany gives the strategic significance of east-data-west-computing project engineering.
With the rapid development of big data, internet, information technology and artificial intelligence technology, cloud computing technology has received great attention in the field of medical and health care at home and abroad. The application of cloud computing technology in the intelligent medical integration platform not only improves the security of medical information, but also reduces the operation and maintenance cost of the intelligent medical platform, which greatly promotes the development of intelligent medical treatment. This article first summarizes the development and applications of cloud computing and intelligent medical treatment at home and abroad in recent years. Then it points out the key problems and solutions in the construction process of intelligent medicine industry, as well as the future development direction of cloud computing technology in intelligent medical treatment.