Buffer Node Queue Graph Configuration for Packet Prioritization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current buffer management systems in communications networks lack flexibility and effectiveness in prioritizing packets, especially for non-IP traffic and encrypted/VPN traffic, as they rely on implementation-specific queueing methods that do not account for varying application requirements and external signaling dependencies.
Innovation Solution
A method and apparatus for buffer management that utilize a complex queue configuration with a queue graph configuration, including packet classifiers, multiple queues, and dequeuers, allowing for dynamic classification and prioritization of packets based on classification types, enabling configurable and adaptable queueing strategies.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If a simple queueing method (FIFO, LIFO) is used, then the device complexity is reduced, but the adaptability to different application requirements deteriorates
Solution Approach 1:
The patent implements dynamic queueing by allowing the queueing method to be selected and configured based on application requirements. The system can dynamically switch between different queueing methods (FIFO, LIFO, priority queues, weighted fair queuing) depending on the traffic flow characteristics and application needs, making the system adaptable without requiring complex hard-coded logic for each scenario
Solution Approach 2:
The patent changes the parameter of queueing method selection from fixed to configurable. By introducing configuration files that define queueing parameters and methods, the system can adjust its behavior to match different application requirements while maintaining a relatively simple core queueing structure
2Measurement precision
If deep packet inspection is performed to perfectly classify packets, then the measurement precision of packet classification is improved, but the use of energy and processing resources increases
Solution Approach 1:
The patent applies partial packet inspection by examining only the necessary packet headers and configuration information to determine queueing method selection, rather than performing complete deep packet inspection on all packets. The configuration files pre-defined queueing rules, so the system only needs to match packets against these pre-established criteria, reducing processing overhead while maintaining classification accuracy
3Adaptability or versatility
If multiple queueing methods are implemented to handle different traffic types, then the adaptability to diverse traffic is improved, but the device complexity increases
Solution Approach 1:
The patent creates a universal queueing framework that can handle multiple queueing methods through a common configuration-driven architecture. Instead of implementing separate complex systems for each queueing method, the patent uses a single configurable system that can switch between FIFO, LIFO, priority queues, and weighted fair queuing based on configuration files, reducing overall system complexity while maintaining versatility
Data Source
AI summary
A buffer node and method of buffering packet data units are provided. The buffer node comprises a receive interface configured to receive a complex queue configuration and to receive packet data units, a buffer configured to implement a queue graph configuration associated with the complex queue configuration to buffer incoming packets into queues, and a send interface configured to send the dequeued incoming packets to a destination node. The queue graph configuration comprises at least one classifier configured to determine a classification type for an incoming packet, at least two queues configured to store incoming packets, and at least one dequeuer configured to select a queue from the at least two queues to be dequeued and to apply a dequeueing function on that selected queue. The method comprises a buffer node receiving a complex queue configuration from a configuration entity, receiving a plurality of packet data units from at least one the packet source node, and sending each incoming packet to one of the at least two queues based on the queue graph configuration. The complex queue configuration is associated with a queue graph configuration for buffering packet data units.