Multipath Adaptive Streaming Controller for Dynamic CDN Load-Balancing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing adaptive HTTP streaming systems lack dynamic load-balancing capabilities between Content Delivery Network (CDN) servers, leading to inconsistent quality and high operational costs, as users are not migrated to better CDN providers during service and load-balancing decisions are coarse, limiting service providers' ability to ensure fair and consistent user experiences.
Innovation Solution
A method and apparatus for controlling data transmission between multiple CDN servers and receivers, using a controller to compute and send control parameters based on real-time information about data reception, such as server identifiers, bit-rates, buffer sizes, and quality indicators, to adjust data requests and optimize bandwidth usage, allowing for dynamic switching between CDN servers and adaptive streaming algorithm parameters.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If load-balancing is performed at client admission time only, then the CDN aggregator can simplify the load-balancing mechanism with coarse policies, but users cannot be migrated to better CDN providers during service, resulting in inconsistent quality experience
Solution Approach 1:
The patent implements dynamic load-balancing by enabling users to switch between CDN providers during the streaming service, not just at admission time. The system continuously monitors service quality and automatically migrates users to better CDN providers, making the load-balancing mechanism adaptive and dynamic rather than static.
Solution Approach 2:
The system employs feedback mechanisms by continuously monitoring service quality metrics during streaming and using this information to make real-time load-balancing decisions. The controller receives quality feedback from ongoing sessions and adjusts CDN selection accordingly, enabling continuous optimization of service quality.
2Loss of energy
If the service provider turns off a CDN provider to reduce operating costs, then operational costs decrease, but currently connected users cannot be migrated to other CDN providers, resulting in service disruption
Solution Approach 1:
The system enables dynamic migration of users between CDN providers during service. When a CDN provider is turned off or becomes unavailable, the system automatically detects this change and migrates affected users to alternative CDN providers, ensuring service continuity while allowing cost optimization through CDN provider management.
Solution Approach 2:
The system performs preliminary actions by pre-establishing multipath streaming capabilities and maintaining alternative CDN provider connections. This allows seamless migration to backup CDN providers when the primary one is shut down, ensuring service continuity without disruption to users.
3Ease of operation
If coarse load-balancing policies are used, then the load-balancing mechanism can be simpler to implement, but the service provider's ability to ensure fair and consistent user experience is significantly reduced
Solution Approach 1:
The system transitions from coarse, static load-balancing policies to fine-grained, dynamic control. The controller continuously adjusts CDN selection and resource allocation based on real-time service quality metrics, enabling precise control over service quality while maintaining ease of operation through automated decision-making.
Solution Approach 2:
The system uses feedback from continuous quality monitoring to refine load-balancing decisions in real-time. This enables the service provider to maintain simple policy management while achieving high precision in service quality control through automated, data-driven adjustments to CDN selection and resource allocation.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
The present invention concerns a method to control the transmission and transmission data rate between at least two servers (1,2) and a receiver (4), the servers (1,2) being adapted to transmit data representative of an audiovisual content, the audiovisual content being available from each of said at least two servers (1,2) in at least two versions, the versions corresponding respectively to different transmission bit-rates, the servers (1,2) being adapted to transmit audiovisual contents in successive parts, each of said successive parts being chosen as one of at least two versions in response to transmission requests sent by the receivers (4), said transmission requests comprising a transmission parameter, the method comprising the steps of reception by a controller (6) of information from the receiver (4), the transmission of at least a control parameter from the controller (6) to the receivers (4), the at least one control parameter being defined from the information in order to define the transmission parameters of the requests from the receiver (4) to one of at least two servers (1,2) or a reception parameter of the receiver (4).