Weighted Arbitration for NoC QoS Fairness
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In Network on Chip (NoC) systems, traditional interconnect architectures face challenges in providing fair and weighted allocation of bandwidth to various components, leading to unfair resource distribution due to scalability limitations and congestion, especially in complex multi-core systems with heterogeneous topologies.
Innovation Solution
The method involves computing weights for channels based on bandwidth requirements and dynamically adjusting these weights to perform weighted arbitration, ensuring Quality of Service (QoS) and fair allocation of resources by monitoring flow activity across the NoC, using a weight allocation module and a weighted arbitration module to implement weight round-robin arbitration.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional interconnect architectures are used, then the system can maintain simplicity in arbitration, but unfair resource distribution occurs due to scalability limitations and congestion
Solution Approach 1:
The patent changes the arbitration parameter from simple round-robin or priority-based selection to weight-based selection. Each channel is assigned a weight value that reflects its bandwidth requirements and QoS needs. The arbitration mechanism selects channels with higher weights preferentially, enabling fair resource distribution while maintaining scalability in NoC systems.
Solution Approach 2:
The patent implements dynamic weight adjustment where channel weights are not fixed but can be modified based on current network conditions, traffic patterns, and QoS requirements. This dynamic adaptation allows the arbitration mechanism to respond to changing system states and maintain fairness under varying loads.
2Adaptability or versatility
If uniform arbitration is used, then the arbitration mechanism remains simple, but bandwidth allocation does not meet diverse QoS requirements of different flows
Solution Approach 1:
The patent applies local quality by assigning different weight values to different channels based on their specific QoS requirements and traffic characteristics. Instead of uniform treatment, each channel receives customized weight assignment that reflects its local needs, enabling differentiated QoS support across the network.
Solution Approach 2:
The patent performs preliminary weight computation based on flow bandwidth requirements and QoS parameters before arbitration occurs. This pre-computation of weights allows the arbitration mechanism to directly use these pre-calculated values without complex real-time calculations, reducing arbitration logic complexity while maintaining QoS adaptability.
3Productivity
If static weight assignment is used, then the arbitration logic is simpler, but unfair allocations occur when traffic conditions change
Solution Approach 1:
The patent implements feedback mechanisms where the arbitration mechanism monitors network traffic conditions, channel utilization, and QoS metric violations. Based on this feedback, weight values are dynamically adjusted to correct unfair allocations and optimize system performance. The feedback loop ensures the system adapts to changing traffic patterns while maintaining QoS guarantees.
Data Source
AI summary
Systems and methods described herein are directed to solutions for NoC interconnects that provide end-to-end uniform- and weighted-fair allocation of resource bandwidths among various contenders. The example implementations are fully distributed and involve computing weights for various channels in a network on chip (NoC) based on the bandwidth requirements of flows at the channels. Example implementations may involve using the weights to perform weighted arbitration between channels in the NoC to provide quality of service (QoS). The weights may be adjusted dynamically by monitoring the activity of flows at the channels. The newly adjusted weights can be used to perform the weighted arbitrations to avoid unfair bandwidth allocations.