Multipath Traffic Steering With Transport Conversion for Legacy Endpoints
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Network operators face challenges in managing multipath connections due to endpoint support limitations and inefficient path selection, leading to unnecessary use of high-cost paths or paths with different subscription requirements, which affect throughput and signaling.
Innovation Solution
Implementing a transport converter that provides backwards compatibility for endpoints lacking multipath support, along with a policy-based traffic steering mechanism that determines path quality using parameters like throughput, round trip time, and packet loss ratio to optimize data transmission.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If a multipath connection is established to enable data transmission over multiple paths, then network reliability and throughput are improved, but device complexity increases due to the need for transport converters and policy management mechanisms
Solution Approach 1:
A transport converter is introduced as an intermediary component that sits between the endpoint and the multipath network infrastructure. This converter translates single-path communication from legacy endpoints into multipath communication protocols, enabling backward compatibility without requiring modifications to existing endpoints. The transport converter handles path selection, packet reassembly, and protocol translation, thereby improving network reliability while isolating the complexity from endpoints.
Solution Approach 2:
The multipath communication system is segmented into distinct functional components: endpoint devices, transport converters, network nodes, and policy management systems. Each component has a specific responsibility - endpoints generate traffic, transport converters handle path translation, network nodes route packets, and policy systems manage path selection criteria. This segmentation allows the system to achieve high reliability through distributed path management while containing device complexity within specialized components rather than requiring all devices to be fully multipath-capable.
2Measurement precision
If path quality is determined using multiple parameters (throughput, round trip time, packet loss), then path selection accuracy is improved, but measurement precision requirements increase
Solution Approach 1:
The system performs preliminary path characterization by continuously monitoring network parameters (throughput, round trip time, packet loss) and storing baseline performance data for each available path. This preliminary measurement phase allows the policy system to pre-evaluate path quality before actual data transmission begins, reducing the need for complex real-time measurements during active communication. The pre-collected measurement data is used to make informed path selection decisions, thereby improving accuracy while reducing the burden of continuous precise measurement.
Solution Approach 2:
A feedback mechanism is implemented where the system continuously monitors actual path performance parameters (throughput, latency, packet loss) and compares them against expected values. This feedback loop allows the policy system to dynamically adjust path selection criteria and update path quality assessments in real-time. The feedback mechanism uses simplified threshold-based comparisons rather than requiring ultra-precise measurements, achieving good path selection accuracy through iterative refinement rather than single-shot precise measurement.
3Productivity
If throughput-based path selection is implemented to optimize data transmission, then productivity is improved, but loss of information increases due to potential packet loss on selected paths
Solution Approach 1:
The path selection mechanism is designed to be dynamic rather than static. The policy system continuously monitors throughput and packet loss parameters for all available paths and adjusts path selection in real-time based on current network conditions. When a high-throughput path begins experiencing packet loss, the system dynamically switches to an alternative path that maintains both throughput and reliability. This dynamic adaptation allows the system to maximize productivity while minimizing information loss by responding to changing network conditions rather than committing to a single path selection.
Solution Approach 2:
The system employs partial path utilization by selecting and actively using only the subset of paths that currently meet both throughput and packet loss criteria, rather than attempting to utilize all available paths simultaneously. The policy system evaluates multiple paths but activates only those that satisfy the dual constraints of adequate throughput and acceptable packet loss rates. This partial action approach ensures productivity goals are met through optimal path selection while inherently protecting against information loss by excluding problematic paths from active use.
Data Source
AI summary
Disclosed are improvements in communication methods and techniques where more than one communication path is available. The methods include determining a first throughput of the first path. The first throughput may be based on a first path of a connection, a mode selection, and analysis of a first packet configured to determine a parameter of the first path. The methods may include determining a second throughput, and the methods may further include sending a third packet comprising data of an application. The third packet may be based on the first throughput and the second throughput.