Asymmetric Relay Node Selection for Overlay Network Latency
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing communication systems face challenges in optimizing relay node selection for bi-directional communication sessions over overlay networks, leading to asymmetrical latencies and suboptimal user experiences due to the reliance on a single relay node for both directions of data transfer.
Innovation Solution
The method involves selecting different relay nodes for each direction of data transfer in a bi-directional communication session based on performance measurements such as latency, bandwidth, and packet loss, allowing for independent optimization of routes for each direction to minimize latency and enhance user experience.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If a single relay node is used for both directions of data transfer, then the system complexity is reduced and ease of operation is improved, but asymmetrical latencies occur and overall performance is suboptimal
Solution Approach 1:
The patent segments the relay node selection process by direction of data transfer. Instead of selecting a single relay node for bidirectional communication, the system independently selects relay nodes for each direction (e.g., relay node A for client-to-server traffic, relay node B for server-to-client traffic). This segmentation allows optimization of latency for each direction separately while maintaining operational simplicity through automated selection based on performance measurements.
Solution Approach 2:
The patent implements dynamic relay node selection where the system continuously measures performance metrics (latency, bandwidth, packet loss) and adjusts relay node assignments based on current network conditions. This dynamic approach allows the system to adapt to changing network states, selecting the most appropriate relay nodes for each direction at any given time, thereby minimizing latency while maintaining ease of operation through automated decision-making.
2Loss of time
If different relay nodes are selected for each direction of data transfer, then latency is minimized and performance is optimized, but the device complexity increases
Solution Approach 1:
The patent implements a self-service mechanism where the routing system automatically performs performance measurements, analyzes results, and selects appropriate relay nodes for each direction without requiring manual configuration or complex centralized control. The system serves itself by continuously monitoring network conditions and autonomously adjusting relay node assignments, thereby optimizing latency while avoiding the complexity of manual management or complex coordination protocols.
Solution Approach 2:
The patent employs feedback mechanisms where performance measurements (latency, bandwidth, packet loss) are continuously collected from actual data transfers, and this feedback is used to inform subsequent relay node selection decisions. The system uses this feedback loop to dynamically adjust relay node assignments, ensuring optimal performance while managing complexity through data-driven automated decision-making rather than complex predetermined routing logic.
3Reliability
If relay nodes are selected based on comprehensive performance measurements, then the quality of service is improved, but the measurement and detection difficulty increases
Solution Approach 1:
The patent applies partial action by selecting a subset of the most critical performance metrics (latency, bandwidth, packet loss) rather than attempting to measure and analyze all possible network parameters. This focused measurement approach achieves sufficient quality of service optimization without the excessive complexity of comprehensive network characterization, allowing the system to make effective relay node selection decisions based on the most impactful performance indicators.
Data Source
Figure 1
Figure 2
AI summary
Routing data in a bi-directional communication session over an overlay network is described. One-way performance measurements are determined for routing data in the communication session in a first direction from a first node to a second node via a respective plurality of relay nodes of the overlay network. Based on the performance measurements, relay node(s) are selected for use in routing data in the first direction from first node to second node. Data is routed in the communication session from the first node to the second node over the overlay network via the selected relay node(s). The selection of the relay node(s) for use in routing data in the first direction is performed separately to selection of one or more relay node for use in routing data in a second direction from the second node to the first node in the communication session.