NDN Forwarding Daemon Congestion Control
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing network congestion control mechanisms, such as TCP congestion control, have become ineffective due to the dynamic nature of Internet traffic, leading to issues like transmission delay and packet loss.
Innovation Solution
Integration of Named Data Networking (NDN) Forwarding Daemon (NFD) into TCP/IP networks to manage congestion by processing Interest and Data packets, using local data stores and routing tables to optimize packet forwarding and reduce traffic overload.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If TCP congestion control is used to manage network congestion, then congestion avoidance is achieved through congestion window management, but transmission delay and packet loss increase under dynamic Internet traffic conditions
Solution Approach 1:
The patent segments the congestion control function into multiple independent components: Interest packet processing, Data packet processing, and congestion detection. Each component operates independently with specific logic for handling different packet types, allowing optimized control without the delays inherent in TCP's monolithic approach
Solution Approach 2:
The system performs preliminary actions by processing Interest packets before Data packets arrive. By maintaining a Pending Interest Table and detecting congestion conditions in advance through Interest packet analysis, the system can prepare buffering decisions and routing paths before actual data transmission begins, reducing transmission delay
2Reliability
If TCP congestion control mechanisms are used, then network congestion is addressed through standard protocols, but packet loss increases due to ineffective congestion management
Solution Approach 1:
The patent implements feedback mechanisms where the system continuously monitors network conditions through Interest packet processing and Pending Interest Table updates. When congestion is detected via feedback from packet arrival patterns and link status, the system adjusts Data packet forwarding decisions in real-time, reducing packet loss by preventing overload before it occurs
Solution Approach 2:
The system introduces an intermediary layer between the data source and network infrastructure through the Named Data Networking Forwarding Daemon. This intermediary processes Interest packets and manages buffering decisions independently of traditional TCP congestion control, effectively mediating congestion management to reduce packet loss
3Productivity
If traditional congestion control mechanisms are applied, then network infrastructure is used for traffic management, but device complexity increases due to dynamic traffic handling requirements
Solution Approach 1:
The patent applies multi-functionality by having the Named Data Networking Forwarding Daemon handle multiple tasks: Interest packet processing, Data packet forwarding, congestion detection, and buffering management. This universal approach consolidates multiple functions into a single system component, reducing overall device complexity while maintaining high traffic management capability
Solution Approach 2:
The system performs self-service by autonomously processing Interest packets and making buffering decisions based on its own internal state (Pending Interest Table and congestion detection). This self-contained approach eliminates the need for complex external congestion control mechanisms, reducing device complexity while maintaining productivity
Data Source
AI summary
Technologies are described to control network congestion in packet-based networks. In some examples, a method may include receiving an Interest packet requesting content, and returning the content from a local data store if the content is in the local data store. The method may also include determining whether a previous request has been made for the requested content if the content is not in the local data store, and creating a record of the Interest packet and discarding the Interest packet if a previous request has been made. The method may further include determining whether a local IP routing table includes an entry that matches a destination IP address specified by the Interest packet if a previous request has not been made, and forwarding the Interest packet if the destination IP address is in the local IP routing table.


