Congestion Control in Caching Video Networks
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional congestion control methods in media delivery networks, such as admission control and reducing video quality, are inefficient in caching and buffering-based systems, leading to unnecessary user frustration and revenue loss, as they fail to effectively manage network congestion.
Innovation Solution
A method that monitors congestion and adapts the level of media personalization in media delivery networks by prioritizing high-popularity media assets during congestion, reducing the number of distinct media assets delivered to decrease network load, and adjusting personalization levels based on congestion levels to optimize cache reuse and bandwidth usage.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If admission control is applied to deny new video deliveries in congested networks, then quality deterioration of established flows is avoided, but unnecessary user frustration and revenue loss occur when local cache could have served the request
Solution Approach 1:
The system implements feedback mechanisms by monitoring cache hit ratios and delivery requests in real-time. When congestion is detected, the system adjusts admission control decisions based on feedback about local cache performance, allowing requests that can be served from local cache while denying only those requiring remote server access during congested periods.
Solution Approach 2:
The admission control mechanism transitions from static to dynamic by continuously adjusting its behavior based on network congestion levels and cache performance. The system dynamically modifies acceptance criteria for new deliveries, becoming more permissive when caches are effective and more restrictive when congestion occurs, rather than maintaining fixed control rules.
2Loss of energy
If video quality is reduced to relieve network congestion, then bandwidth requirements are decreased, but additional stress is placed on storage and network due to differently encoded lower-quality video
Solution Approach 1:
Instead of re-encoding video at lower quality, the system creates and distributes copies of video segments to local caches. These cached copies are then delivered to users during congested periods, maintaining original video quality while reducing real-time encoding requirements and network stress from multiple encoding operations.
Solution Approach 2:
The system performs preliminary actions by pre-caching video segments at local nodes before congestion occurs. This advance preparation allows the system to serve users from local caches during congested periods without requiring real-time encoding or high-bandwidth transmission, thereby reducing both bandwidth consumption and storage/network stress.
3Adaptability or versatility
If high level of personalization is applied to deliver both high and low popularity media assets, then user experience is improved, but network load and cache requirements increase
Solution Approach 1:
The system applies local quality by allowing high personalization at edge cache nodes while maintaining reduced personalization at the core network level. Local caches can store and deliver both high and low popularity content to individual users, while the core network only needs to deliver content to caches, thereby reducing overall network load while maintaining user experience quality.
4Device complexity
If conventional admission control is used in caching networks, then simple control mechanisms are maintained, but efficiency is lost by denying requests that could be served from local cache
Solution Approach 1:
The system introduces an intermediary layer between the simple admission control mechanism and the actual delivery decision. This intermediary evaluates cache status and request characteristics, allowing simple control mechanisms to remain at the core while improving efficiency through intelligent intermediation that determines whether requests should be fulfilled from local cache or denied based on congestion conditions.
Data Source
AI summary
The present invention relates to a method, related system, related congestion monitoring device and related media delivery personalization devices for controlling congestion in a media delivery network that includes a media delivery server and a plurality of client devices. The media delivery server is coupled over the media delivery network to each client device of the plurality of client devices. The media delivery server delivering media to a client device over a path between the media delivery server and the client device. The related method includes the steps of monitoring congestion in the media delivery network and subsequently adapting a level of personalization of the delivering of media to the client device based on the congestion monitored.