Dynamic Vector-Based Packet Processing in 5G UPF
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current 5G communication systems and IoT networks face inefficiencies in processing user plane function (UPF) packets, as existing methods do not consider varying packet types, communication protocols, or real-time resource changes, leading to suboptimal processing and resource allocation.
Innovation Solution
A method and apparatus that analyze the packet processing state to determine the size of a packet vector and allocate processing cores to each packet processing pipeline based on real-time conditions, using AI and machine learning to optimize vector size and core allocation for efficient UPF packet processing.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If static resource allocation is used for packet processing pipelines, then device complexity is reduced, but productivity decreases due to inability to adapt to varying packet types and real-time resource changes
Solution Approach 1:
The patent implements dynamic resource allocation where the network processor dynamically determines vector sizes and allocates processing cores to packet processing pipelines based on real-time packet processing states. This allows the system to adapt to varying packet types, protocols, and workload conditions, thereby improving processing throughput without requiring overly complex static configuration mechanisms
Solution Approach 2:
The system changes key parameters (vector size, number of processing cores allocated to each pipeline) based on analyzed packet processing states. By adjusting these parameters dynamically according to real-time conditions such as packet types and protocol distributions, the system optimizes processing efficiency while maintaining manageable complexity through algorithmic control
2Productivity
If vector size is increased for packet processing, then productivity improves through batch processing, but loss of time increases due to longer processing cycles for larger vectors
Solution Approach 1:
The network processor dynamically determines the size of packet vectors based on real-time packet processing states. This allows the system to adjust vector sizes adaptively - using larger vectors when workload permits to improve batch processing efficiency, and reducing vector sizes when processing delays need to be minimized, thereby balancing throughput and latency requirements
Solution Approach 2:
The system changes the vector size parameter based on analyzed processing states including packet types and current pipeline performance. By adjusting this parameter dynamically, the system optimizes the trade-off between batch processing efficiency (larger vectors) and processing cycle time (smaller vectors) according to real-time conditions
3Productivity
If more processing cores are allocated to packet processing pipelines, then productivity increases, but device complexity and resource management difficulty increase
Solution Approach 1:
The network processor automatically analyzes packet processing states and performs dynamic allocation of processing cores to different packet processing pipelines without external intervention. This self-service mechanism manages the complexity of coordinating multiple cores by embedding the allocation logic within the processor itself, simplifying resource management while maximizing processing capacity
Solution Approach 2:
The system dynamically changes the number of processing cores allocated to each pipeline based on real-time analysis of packet types, protocols, and processing states. This parameter adjustment allows the system to scale processing capacity according to actual workload demands while maintaining manageable complexity through automated decision-making algorithms
Data Source
AI summary
Disclosed is a user plane function (UPF) packet processing control apparatus including: a processor, and a memory storing instructions, wherein the processor is configured to execute the instructions to analyze a state in which UPF packets are processed by a plurality of packet processing pipelines, determine a size of a packet vector for each of the plurality of packet processing pipelines based on a result of the analyzing, and allocate a number of processing cores to each of the plurality of packet processing pipelines based on the result of the analyzing.