Dynamic Network Link Weighting and Scheduling
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing network communication systems face inefficiencies when managing multiple network links with varying characteristics such as RTT latency, packet drops, and throughput, leading to suboptimal performance in applications like data replication systems.
Innovation Solution
A system comprising a link mapper, a link weighting component, and a link scheduler that identifies and weights network links based on specific characteristics like MTU, latency, and throughput, and schedules their use to optimize performance.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If arbitrary switching between network links is used, then device complexity is reduced, but network link performance and reliability deteriorate
Solution Approach 1:
The patent applies parameter changes by dynamically adjusting the weight of each network link based on measured performance parameters such as RTT latency, packet drops, timeouts, and throughput. The link weighting component continuously monitors these parameters and recalculates weights to reflect current link quality, allowing the system to adapt to changing network conditions without increasing operational complexity.
Solution Approach 2:
The patent implements feedback mechanisms where the link weighting component receives performance data from the network links, processes this information to determine updated weights, and feeds these weights back to the link scheduler. This closed-loop feedback system enables automatic adaptation to network conditions, improving reliability while maintaining simple arbitrary switching behavior at the operational level.
2Ease of operation
If arbitrary switching between network links is used, then ease of operation is improved, but network link performance deteriorates
Solution Approach 1:
The patent applies self-service by enabling the network link management system to automatically evaluate, weight, and schedule links without manual intervention. The link weighting component and link scheduler work autonomously to optimize link selection based on current performance characteristics, maintaining ease of operation while significantly improving network throughput and application performance.
Solution Approach 2:
The system dynamically changes scheduling parameters based on measured link performance. The link scheduler uses weight parameters derived from real-time monitoring of throughput, latency, and error rates to automatically adjust link selection, achieving high productivity without requiring manual configuration or complex operational decisions.
3Adaptability or versatility
If network links with varying characteristics are used, then adaptability is improved, but performance optimization becomes difficult
Solution Approach 1:
The patent applies local quality by evaluating and weighting each network link individually based on its specific characteristics such as RTT latency, packet drop rate, timeout frequency, and throughput capacity. Instead of treating all links uniformly, the system assigns different weights to different links according to their local performance qualities, enabling optimal utilization of diverse link types including broadband, LTE, and Wi-Fi connections.
Solution Approach 2:
The system changes scheduling parameters dynamically based on the specific characteristics of available links. The link weighting component calculates weights using multiple performance parameters simultaneously, and the link scheduler adjusts link selection based on these composite weight values, achieving performance optimization across diverse link types without requiring manual configuration for each link type.
Data Source
AI summary
The described technology is generally directed towards evaluating, weighting, and scheduling the use of network links. According to an embodiment, a system can comprise a memory that can store computer executable components, and a processor that can execute the computer executable components stored in the memory. The components can comprise a link mapper that can identify at least two network links between a first device and a second device, resulting in identified network links. The system can further comprise a link weighting component that can assign weights to respective ones of the identified network links resulting in weighted network links, based on a criterion and characteristics of the identified network links, the first device and the second device. The system can further comprise a link scheduler that can schedule the weighted network links based on the weights assigned to the respective ones of the weighted network links.