1. The State Key Laboratory of Management and Control for Complex Systems, Institute of Automation, Chinese Academy of Sciences, Beijing 100190, China
2. Research Center of Computational Experiments and Parallel Systems, National University of Defense Technology, Changsha 410073, China
3. School of Automation, Huazhong University of Science and Technology, Wuhan 430074, China
4. School of Electronics Engineering and Computer Science, Peking University, Beijing 100871, China
5. Qingdao Academy of Intelligent Industries, Qingdao 266000, China
Corresponding author (email@example.com
The coexistence of various protocols for current networks leads to extremely complex network systems, which not only limits the development of network technologies, but also cannot meet the growing demands for cloud computing, big data, service visualization applications, etc. With the development of information telecommunication technology, decreasing hardware costs, and the rise of open source packages, network systems have become more dynamic and flexible, and network service providers expect easier access to the information technology market. A new business model and collaborative competition must be formed.However, current network systems are becoming increasingly complex, which also increases project and societal complexity. The attributes of multi-field, dynamicity, and unpredictability lead network systems to be massively complex, making it difficult to comprehensively evaluate and accurately amend their schema. Therefore, as a new network architecture, parallel network is expected to revolutionize the current network situation and meet the evolving demand for network services. The core idea of parallel networking is to construct artificial networks and effectively optimize the network system operations via the interactions between real networks and artificial networks. Through computational experiments and analysis of the artificial networks, a control strategy based on network traffic flow can be continuously tracked and updated in real-time. Meanwhile, the collected operating status of the real network can also be used to optimize the model of the artificial network. These strategies can be applied to all types of network equipment to control network operations. Therefore, it is possible to allocate network resources more effectively, to improve the management and utilization of resources, and to provide new network solutions to address changing network demands more effectively.
Copyright 2019 Science China Press Co., Ltd. 《中国科学》杂志社有限责任公司 版权所有