Media Packet Metadata for Latency Monitoring in Distributed Cloud
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In distributed cloud environments, media processing across multiple virtualized media servers introduces significant latency issues, exceeding acceptable delays for real-time communications like video conferencing, which can lead to user dissatisfaction and service quality degradation.
Innovation Solution
The method involves attaching metadata to media packets that includes processing delay information, allowing for real-time monitoring and adjustment of the media processing chain to maintain acceptable latency levels, ensuring user experience and Service Level Agreements (SLAs) are met by dynamically reconfiguring the media server chain and optimizing processing paths.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If media processing is distributed to multiple virtual machines in different data centers, then responsiveness and user experience are improved, but end-to-end latency increases
Solution Approach 1:
The system performs preliminary actions by pre-calculating and pre-positioning media packets at intermediate nodes before they are actually needed. This allows the media server to retrieve pre-positioned packets rather than generating them in real-time, significantly reducing end-to-end latency while maintaining the benefits of distributed processing.
Solution Approach 2:
The invention introduces intermediary components including a media controller that coordinates between multiple media servers, and intermediate nodes that store pre-positioned packets. These intermediaries enable efficient routing and retrieval of media content across the distributed system, reducing the time penalty associated with distributed architecture.
2Loss of time
If media processing is centralized in a single high-capacity data center, then end-to-end latency is reduced, but responsiveness and adaptability to user feedback decrease
Solution Approach 1:
The system segments media processing functions across multiple media servers distributed in different data centers, with each server handling specific media streams or user groups. This segmentation enables localized processing that improves responsiveness to user feedback while the coordination layer maintains low latency through intelligent routing and pre-positioning strategies.
Solution Approach 2:
The system implements dynamic routing and adaptive pre-positioning where media packets are routed through different paths based on real-time network conditions and user feedback. The media controller dynamically adjusts which intermediate nodes store and forward packets, enabling the system to adapt to changing conditions while maintaining low latency performance.
3Productivity
If multiple media servers are used in the media path, then processing capacity is increased, but delay budget is exceeded
Solution Approach 1:
The system performs preliminary actions by pre-positioning media packets at intermediate nodes before they are needed by the media server. This eliminates the need for real-time generation and forwarding through multiple servers, maintaining high processing capacity while reducing the cumulative delay that would otherwise exceed the delay budget.
Solution Approach 2:
The invention enables the media server to skip the traditional time-consuming process of real-time media packet generation and forwarding by directly retrieving pre-positioned packets from intermediate nodes. This rushing through of the processing chain maintains high productivity while minimizing the time spent in the media path, keeping delay within budget.
Data Source
AI summary
The disclosure relates to a method 30 for handling latency performed in a node 12, 151, 152, 153, . . . , 15n of a distributed cloud 11. The node 12, 151, 152, 153, . . . , 15n is configured to handle a subset of media processing required by a media service and the distributed cloud 11 comprising two or more such nodes 12, 151, 152, 153, . . . , 15n, wherein a communication path for the media service is configured between at least a first and a second communication device 141, 142. The media service is provided by media streams comprising media packets and the communication path involving two or more of the nodes 12, 151, 152, 153, . . . , 15n. The method 30 comprises: determining 31 processing delay of processing the media packet in the node 12, 151, 152, 153, . . . , 15n the processing delay constituting a first delay contribution; attaching 32 metadata to the media packet, the metadata comprising information about the first delay contribution added to any accumulated delay contributions experienced by the media packet along the communication path.