Application-Aware Packet Routing for Fine-Grained Traffic Scheduling
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current traffic scheduling in networks is based on quintuple information, leading to a coarse management granularity and poor allocation rationality of network resources, as packets with the same quintuple information often have varying traffic requirements.
Innovation Solution
Implement traffic scheduling based on application program-level management by encapsulating feature information, such as bandwidth, latency, and security requirements, into IPv6 extension headers or destination addresses, allowing nodes to determine optimal paths and queues for packet forwarding.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If traffic scheduling is performed based on quintuple information, then the management process is simple, but the management granularity is coarse and allocation rationality is poor
Solution Approach 1:
The patent segments the traffic scheduling management by introducing application program-level granularity. Instead of treating all packets with the same quintuple information uniformly, the system divides traffic management into finer segments by identifying and managing packets based on their application program characteristics. This segmentation allows different application programs to receive differentiated traffic scheduling treatments, thereby improving management granularity while maintaining operational simplicity through automated application identification mechanisms.
2Ease of manufacture
If traffic scheduling is performed based on quintuple information, then the implementation is straightforward, but the allocation rationality of network resources is poor
Solution Approach 1:
The patent applies local quality by enabling different parts of the network traffic to have different management characteristics based on their application program identity. Each application program can be assigned specific quality of service parameters, bandwidth limits, and scheduling priorities tailored to its requirements. This allows the network to adapt resource allocation to local needs of different applications while maintaining a unified implementation framework that builds upon existing quintuple-based routing infrastructure.
3Productivity
If packets with same quintuple information are treated uniformly, then the processing is efficient, but the network resource allocation rationality deteriorates
Solution Approach 1:
The patent implements preliminary action by pre-identifying and tagging packets with their application program information before they enter the core routing process. Application identification and classification are performed in advance, allowing the network to make informed scheduling decisions without adding significant processing overhead during packet forwarding. This preliminary classification enables differentiated resource allocation while maintaining high processing efficiency through optimized packet handling paths.
Data Source
AI summary
An ingress node obtains a packet, and feature information of an application program corresponding to the packet. The feature information of the application program includes at least one of the following information: identification information and network performance requirement information of the application program. The identification information is used to identify the application program. The ingress node determines the path between the ingress node and the egress node based on the feature information of the application program and a mapping relationship between the feature information of the application program and the path between the ingress node and the egress node. The ingress mode forwards the packet through the path between the ingress node and the egress node.