QoS-Based Core Mapping for Low-Latency Packet Processing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing electronic devices face latency issues in processing data packets due to the application processor's inability to handle packets with different quality of service (QoS) requirements, particularly for low-latency services like URLLC, as they are delayed by normal packets during merging.
Innovation Solution
The application processor shares group information based on QoS characteristics with the communication processor, mapping packets to corresponding cores for parallel processing, reducing latency by configuring sessions to specific cores based on QoS.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of time
If packets are merged in the order received from the communication processor, then packet processing follows a simple sequential approach, but low-latency service packets (e.g., URLLC) are delayed by normal packets
Solution Approach 1:
The patent segments packets into different groups based on QoS characteristics and processes them through different cores. Low-latency service packets are separated from normal packets and handled by dedicated cores, eliminating delays caused by sequential processing of all packets together.
Solution Approach 2:
The application processor preliminarily configures group information and mapping relationships between packet groups and cores before packet processing begins. This pre-configuration enables immediate routing of packets to appropriate cores based on their QoS requirements, reducing processing latency without requiring complex real-time decisions.
2Productivity
If multiple packets are processed through a single control core, then the processing logic is simple, but the processing speed and throughput are limited
Solution Approach 1:
The patent merges multiple packet processing functions across multiple cores, allowing parallel processing of different packet groups. By combining the processing capabilities of multiple cores and assigning different packet groups to different cores, the system achieves higher throughput while maintaining manageable complexity through systematic packet distribution.
Solution Approach 2:
The patent introduces a new dimension of parallel processing by utilizing multiple cores simultaneously. Instead of processing packets sequentially through a single core, the system distributes packets across multiple cores in parallel, significantly increasing processing throughput while the mapping mechanism keeps coordination complexity manageable.
3Reliability
If all packets are processed through the same core, then resource management is simple, but QoS requirements of different services cannot be met
Solution Approach 1:
The patent applies local quality by assigning specific packet groups with particular QoS requirements to specific cores optimized for those requirements. Low-latency service packets are routed to cores configured for fast processing, while normal packets are handled by other cores, ensuring that each packet group receives the appropriate processing quality needed for its service requirements.
Solution Approach 2:
The mapping module provides a universal mechanism that works across different packet groups and service types. By establishing a general mapping relationship between packet groups and cores that can accommodate various QoS requirements, the system achieves reliable QoS guarantees for multiple services while using a unified management approach rather than service-specific configurations.
Data Source
AI summary
An electronic device is provided. The electronic device includes an application processor comprising a plurality of cores, and a communication processor electrically connected to the application processor. The application processor may configure at least one session associated with respective applications based on quality of service (QoS) so as to correspond to at least one core from among the plurality of cores, transmit, to the communication processor, information corresponding to a first core configured for a first session from among the at least one session, receive a data packet of the first session from the communication processor, and process the received data packet of the first session by means of the first core configured for the first session, based on the information corresponding to the first core.