Reflected Packets for Dynamic Network Path Weighting
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Complex computer networks face challenges in optimizing traffic flow due to multiple paths between nodes, which can lead to congestion and inefficiencies, as traditional path selection methods do not adequately consider current network state, such as congestion and path faults.
Innovation Solution
Implementing dynamic weighted cost multipathing and using reflected packets to collect state information, which allows nodes to adjust path weights based on real-time network conditions, and transforming dropped packets into special visibility packets for analysis and correction.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional path selection methods are used, then device complexity is reduced, but network performance and reliability deteriorate due to inadequate consideration of current network state
Solution Approach 1:
The patent implements feedback mechanisms where nodes continuously monitor network state (congestion, faults) and use this information to dynamically adjust path selection. Reflected packets carry state information back to source nodes, enabling closed-loop control that improves reliability while managing complexity through structured feedback channels.
Solution Approach 2:
The patent transitions from static path selection to dynamic path selection where path weights and routing decisions are continuously adjusted based on real-time network state. This dynamic adaptation allows the network to respond to changing conditions, improving reliability without requiring overly complex centralized control.
2Productivity
If multiple paths are used between nodes, then network capacity and redundancy are improved, but congestion and inefficiencies increase due to inadequate path optimization
Solution Approach 1:
The patent changes the parameters of path selection by introducing dynamic weight adjustments based on network state. Instead of using fixed path metrics, the system continuously modifies path weights according to congestion levels, link status, and other state information, enabling optimal utilization of multiple paths while maintaining efficiency.
Solution Approach 2:
The patent segments the network into multiple paths with independently optimized weights, allowing different portions of traffic to be routed through different paths based on current conditions. This segmentation enables parallel optimization of multiple routes rather than treating the network as a single monolithic path.
3Measurement precision
If real-time network state monitoring is implemented, then path optimization is improved, but information collection overhead and processing complexity increase
Solution Approach 1:
The patent makes reflected packets multi-functional by using them both for their primary purpose (delivering data) and for state monitoring (carrying state information). This universal use of existing packet infrastructure enables network state visibility without requiring separate dedicated monitoring channels, reducing information collection overhead.
Solution Approach 2:
The patent implements self-service monitoring where the network infrastructure itself generates and processes state information through reflected packets. Nodes automatically monitor and report their own state without requiring external monitoring systems, reducing the overhead of information collection while maintaining precise measurement capabilities.
Data Source
AI summary
Nodes within a network are configured to adapt to changing path states, due to congestion, node failures, and/or other factors. A node may selectively convey path information and/or other state information to another node by annotating the information into packets it receives from the other node. A node may selectively reflect these annotated packets back to the other node, or other nodes that subsequently receive these annotated packets may reflect them. A weighted cost multipathing selection technique is improved by dynamically adjusting weights of paths in response to feedback indicating the current state of the network topology, such as collected through these reflected packets. In an embodiment, certain packets that would have been dropped may instead be transformed into “special visibility” packets that may be stored and/or sent for analysis. In an embodiment, insight into the performance of a network device is enhanced through the use of programmable visibility engines.