Adaptive Media Pipeline Concurrency Control with Priority Scoring
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The delivery of data streams via a cloud network experiences increased network load during high-demand events, leading to service degradation and component failures, which existing methods address by throttling all traffic equally, causing uniform performance degradation across users.
Innovation Solution
A system utilizing a machine learning model to prioritize data streams based on request data and network status, determining priority scores to differentiate and manage network load by providing high-priority streams at optimal rates while queuing lower-priority streams, and managing retry requests through scheduled intervals.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If all traffic is throttled equally during high network load, then network load is reduced, but service quality deteriorates uniformly for all users
Solution Approach 1:
The patent applies local quality by differentiating traffic treatment based on user priorities. High-priority users receive preferred treatment (lower throttling) while low-priority users receive standard treatment (higher throttling), allowing the system to reduce overall network load while maintaining service quality for important users during high-load conditions.
Solution Approach 2:
The patent segments users into different priority groups (high-priority and low-priority) and applies different throttling strategies to each segment. This segmentation enables the system to manage network load by controlling traffic from different user groups differently, rather than applying uniform throttling to all users.
2Reliability
If high-priority data streams are provided at optimal rates, then service quality for important users is maintained, but overall network load increases
Solution Approach 1:
The patent applies partial action by providing optimal rates only to high-priority data streams while applying reduced rates to low-priority streams. This selective application of optimal service levels ensures that critical users receive full service quality while the system maintains overall load control through differentiated treatment of non-critical users.
3Quantity of substance
If retry requests are managed through scheduled intervals, then network load is reduced, but request response time increases
Solution Approach 1:
The patent applies periodic action by implementing scheduled intervals for retry requests. Instead of allowing continuous or immediate retries, the system spaces out retry attempts at defined intervals, which reduces the overall network load from retry traffic while still enabling users to eventually receive their requested data streams.
Data Source
AI summary
A method may include receiving a network alert associated with a cloud network configured to provide a plurality of data streams to user devices and including one or more network components. The network alert may include request data associated with one or more requests received from respective user devices. The method may include determining a network status of the cloud network indicating a network load. The method may include providing the request data and/or the network status to a machine learning model configured to assign a priority score of each data stream. The method may include determining a high-priority data stream and a second data stream. The method may include providing first instructions indicating that the high-priority data stream is to be provided to a first user device according to respective requests.