Dynamic Traffic Steering System for Video Optimization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Static traffic steering rules become outdated quickly due to the rapid changes in Internet-delivered video content, leading to suboptimal delivery of video data without the benefits of quality of service (QOS) optimization.
Innovation Solution
A dynamic traffic steering system that continuously updates steering criteria based on IP-based learning, using a whitelist, blacklist, and discovery proxy to identify and optimize video content delivery, ensuring that traffic is steered to appropriate optimization resources.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If static steering rules are used, then implementation is simple, but the rules become outdated quickly and fail to optimize video delivery
Solution Approach 1:
The patent implements dynamic traffic steering by continuously updating steering criteria based on IP-based learning from observed traffic patterns. The system transitions from static, pre-configured rules to dynamic rules that adapt automatically to changing video content sources and delivery patterns, resolving the contradiction between adaptability and complexity.
Solution Approach 2:
The system employs feedback mechanisms where traffic steering decisions are continuously refined based on observed traffic patterns and performance metrics. The steering criteria are updated using feedback from actual video delivery traffic, enabling the system to adapt to changing conditions while maintaining manageable complexity through data-driven optimization.
2Productivity
If dynamic traffic steering with continuous updates is implemented, then video delivery optimization is improved, but system complexity and processing overhead increase
Solution Approach 1:
The traffic steering system performs self-updates by automatically learning from observed traffic patterns without requiring manual intervention or complex external management systems. The IP-based learning mechanism enables the system to autonomously identify video content sources and update steering criteria, improving productivity while keeping processing complexity manageable through self-organizing behavior.
Solution Approach 2:
The system performs preliminary learning and pattern recognition on traffic data before making steering decisions. By pre-processing traffic information and establishing patterns in advance, the system prepares optimization rules that can be applied efficiently during actual video delivery, improving productivity without proportionally increasing real-time processing complexity.
3Reliability
If static steering rules are used, then processing overhead is low, but video content may not be delivered to optimization resources
Solution Approach 1:
The patent implements continuous traffic steering optimization by continuously updating steering criteria based on ongoing traffic observation and IP-based learning. This continuous adaptation ensures video content is reliably directed to optimization resources without interruption, eliminating the periodic update delays inherent in static rule systems while maintaining low processing overhead through incremental learning.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A method and system for dynamic traffic steering is described. In one embodiment, a method for dynamic traffic steering involves receiving a request for content at a steering component, comparing information in the request with steering criteria in the steering component, steering the request based on the comparing, and continuously updating the steering criteria based on requests that are subsequently received at the steering component. Other embodiments are also described.