Adaptive traffic management for high-capacity network architectures with smart services

网络服务正在从基于客户机-服务器的服务演变为位置和上下文感知的分布式服务。未来的智能服务将使用上下文和位置动态地向用户设备推送信息。随着这些智能网络服务的快速增长，对网络技术进行调整，使其能够在安全的可扩展平台上提供业务敏捷性是非常重要的。未来互联网的用户将需要一个安全、节能、自管理和可配置的网络平台。我们的研究计划的目标是研究网络流量管理层面的方法和技术，以有效地支持智能服务。最有前途的研究方法之一是采用自主通信进行网络和设备的自配置、自供应和自管理。我们将通过监测网络资源设计动态资源管理方法和算法。随着网络连接速度的不断提高，如何设计出能够满足流媒体音频和视频的高数据速率的路由器和交换设备是一个具有挑战性的问题。我们对这个研究问题的方法是研究可扩展的基于分组的路由/交换体系结构，以有效地传输如此大量的数据。今天的网络大多通过专用的终端系统检测垃圾邮件、病毒和其他恶意网络使用情况。高速骨干路由器通常不会直接使用太多智能来防御这些。我们的研究小组将开发方法和技术，将传输中的无状态设计纳入骨干路由器中，以检测恶意网络使用，作为网络自卫机制。因此，我们的研究将为未来高容量网络架构的关键领域提供高效、安全的方法、算法和技术。这项研究非常及时，对我国具有重要的战略意义。网络中更多的自主性将导致更简单的系统、更好的弹性、更高的稳定性、更高的可用性、更少的错误和更高的安全性。研究网络流量管理层面的方法和技术对未来的智能网络业务至关重要。

Network services are evolving from client-server based services to location and context-aware distributed services. Future smart services will use context and location to push information to user devices dynamically. With the rapid growth of these smart network services, it is important to adapt the network technology so that it can provide service agility over a secure scalable platform. Users of the future Internet will demand a secure, energy-efficient, self-manageable, and configurable network platform. The objective of our research program is to investigate methods and techniques at the network traffic management level to support smart services effectively. One of the most promising research approaches is to employ autonomic communications for network and device self-configuration, self-provisioning, and self-management. We will design dynamic resource management methods and algorithms by monitoring network resources. With ever increasing link speeds, it is a challenging problem to design routers and switching equipment that can cope with the high data rates of streaming audio and video. Our approach to this research problem is to investigate scalable packet based routing/switching architectures to transport such high volumes data effectively. Today's networks mostly detect spam, viruses, and other malicious network usage through dedicated end systems. High-speed backbone routers generally do not directly employ much intelligence to defend these. Our research group will develop methods and techniques to incorporate in-transit stateless designs in the backbone routers to detect malicious network usage as a network self-defence mechanism. Thus, our research will contribute highly efficient and secure methods, algorithms, and techniques to the critical area of future high capacity network architectures. This research is timely and of critical strategic importance for our nation. More autonomicity in networks will lead to simpler systems, better resiliency, more stability, higher availability, fewer errors, and higher security. Investigating methods and techniques at the network traffic management level is critical for the future of smart network service.