On-demand Diverse Path Computation for Limited Visibility Networks
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current diverse path mechanisms in limited visibility computer networks, such as those using Distance Vector routing protocols, rely on full network topology knowledge, which is not available, making it challenging to compute diverse paths efficiently, especially in lossy and resource-constrained Low-Power and Lossy Networks (LLNs) like IoT applications.
Innovation Solution
The solution involves a source device sending a probe packet on a primary path with an indication to transit devices to add their local neighbor topologies, which are then accumulated and used by the destination device to compute a diverse path, allowing on-demand diverse path computation without requiring full topology knowledge.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If full network topology knowledge is used for diverse path computation, then path diversity and reliability are improved, but device complexity and memory requirements increase significantly
Solution Approach 1:
The patent segments the network topology knowledge into local topology information stored at each node. Instead of requiring each node to store the complete network topology, each node only stores and processes its local topology, which is then aggregated along the path. This segmentation dramatically reduces the memory and processing requirements at each node while still enabling diverse path computation.
Solution Approach 2:
The patent introduces probe packets as intermediaries to carry topology information between nodes. These probe packets traverse the network and collect local topology information from intermediate nodes, acting as mediators that enable diverse path computation without requiring each node to maintain a complete topology database. The probe packets facilitate the aggregation of distributed topology information.
2Productivity
If diverse path computation is performed on-demand, then computation efficiency and responsiveness are improved, but processing time and computational overhead increase
Solution Approach 1:
The patent performs preliminary actions by pre-establishing the capability for on-demand diverse path computation and using probe packets to gather topology information in advance. The system is prepared to compute diverse paths when needed, and topology information is collected during normal packet transmission, reducing the need for separate, time-consuming topology discovery processes.
Solution Approach 2:
The patent uses probe packets as copies that traverse the network to collect topology information. These probe packets replicate the functionality of topology discovery without requiring dedicated discovery messages, leveraging existing packet transmission mechanisms to gather information efficiently and reduce computational overhead.
3Reliability
If multi-path forwarding is implemented for critical packets, then reliability and redundancy are improved, but network bandwidth consumption and processing load increase
Solution Approach 1:
The patent applies local quality by implementing multi-path forwarding selectively for critical packets based on their characteristics and requirements. Instead of treating all packets uniformly, the system identifies critical packets (e.g., those requiring redundancy protection) and applies diverse path computation only to those, optimizing bandwidth usage while maintaining reliability where needed.
Solution Approach 2:
The patent changes parameters such as packet priority, type, or marking to identify critical packets that require diverse path forwarding. By modifying packet parameters to indicate their criticality, the system can dynamically adjust forwarding behavior and allocate network resources more efficiently, ensuring redundancy protection for important traffic while minimizing overall bandwidth consumption.
Data Source
AI summary
In one embodiment, a source device detects a packet flow that meets criteria for multi-path forwarding, and forwards a probe packet on a primary path from the source device to a destination device, the probe packet carrying an indication to cause a plurality of transit devices along the primary path to add their respective local neighbor topology to the forwarded probe packet, and also to cause the destination device to compute a diverse path from the primary path based on the accumulated local neighbor topologies in the probe packet. Accordingly, the source device may receive a returned diverse path as computed by the destination device in response to the probe packet, and may thus forward the packet flow on the primary path and the diverse path from the source device to the destination device according to the multi-path forwarding.