QoS Packet Scheduling with Random Priority Keys
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current packet-based telecommunication systems face challenges in managing Quality of Service (QoS) and service provider resources, particularly in satellite systems, where resource allocation is static and not scalable, leading to inefficiencies and unfairness during congestion, and existing Differentiated services algorithms are too complex or unfair.
Innovation Solution
A system that extracts Quality of Service parameters from packet headers, generates random priority keys, and uses programmable rules to schedule packet output, managing queues and resource allocation dynamically to ensure fairness and scalability, while handling capacity overflows and buffering, ensuring negotiated bandwidth and optimal system performance.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If simple Differentiated services algorithms are used, then device complexity is reduced, but fairness during congestion deteriorates
Solution Approach 1:
The patent changes the parameter of priority key generation from static to dynamic by incorporating random values. This allows the scheduling system to maintain simplicity while achieving fairness during congestion through probabilistic priority assignment rather than fixed hierarchical prioritization
Solution Approach 2:
The patent implements feedback mechanisms through random value generation and priority key recalculation. The system continuously adjusts packet priorities based on current congestion conditions and randomization, creating a feedback loop that prevents starvation and ensures fair resource distribution without complex algorithms
2Reliability
If strict provider resource partition is used, then service provider guarantees are maintained, but global system performance deteriorates
Solution Approach 1:
The patent applies dynamics by making the resource partitioning flexible rather than static. The programmable key attribution device allows dynamic adjustment of provider resource allocation based on actual traffic conditions and negotiated parameters, enabling the system to adapt to changing demands while maintaining guarantees
Solution Approach 2:
The patent creates a universal scheduling framework that can handle multiple service providers and QoS classes simultaneously. The single priority key mechanism serves multiple functions: it enforces provider guarantees, optimizes global performance, and accommodates different traffic types without requiring separate rigid partitioning schemes for each provider
3Reliability
If per provider queues are used with fixed partitioning, then service guarantees are ensured, but adaptability to commercial exploitation deteriorates
Solution Approach 1:
The patent transforms the fixed per-provider queue structure into a dynamic programmable key attribution system. The key attribution rules can be programmatically adjusted to accommodate different commercial scenarios, provider agreements, and traffic patterns without reconfiguring the fundamental queue architecture, thereby enhancing adaptability while maintaining service guarantees
Data Source
AI summary
A system for scheduling the attribution of hardware output resources to data packets handled by different service providers is provided. The system includes an extraction device which extracts a Quality of Service parameter included in a packet header; a random value generator; a programmable key attribution device attributing a priority key to a packet depending on the Quality of Service parameter and on the generated random value, according to programmable rules; a scheduling device that includes a command generator generating packet switching commands in an order depending on the value of the priority key attributed to each packet header; and a switching device routing the packets towards hardware output resources in the order of the generated packet switching commands.