Grid Router QoS Enforcement via Distributed Priority Scheduling
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In data networks, especially in congested conditions, routers face challenges in maintaining quality of service (QoS) due to high traffic exceeding bandwidth capacity, leading to packet delays or drops, which affects service quality.
Innovation Solution
A grid router system with distributed units that include mailbox units, egress units, and transport units, equipped with congestion determiners, dynamic priority schedulers, and token bucket shapers, which adjust data flow and priority levels based on network congestion to ensure QoS across links.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If traffic capacity is increased to handle high bandwidth demands, then network throughput is improved, but packet loss and delays increase under congestion
Solution Approach 1:
The network traffic is segmented into different priority classes (high priority and low priority packets). The router maintains separate queues for each priority level, allowing high priority traffic to be transmitted first during congestion events. This segmentation enables the network to handle high throughput while protecting time-sensitive packets from loss and delay.
Solution Approach 2:
Different quality of service treatments are applied to different packets based on their priority classification. High priority packets receive preferential treatment with guaranteed transmission opportunities and lower drop probabilities, while low priority packets receive best-effort service. This local differentiation of service quality allows the system to maintain reliability for critical traffic while maximizing overall throughput.
2Reliability
If strict quality of service guarantees are provided for all packets, then packet delivery reliability is improved, but network complexity increases
Solution Approach 1:
Instead of providing strict QoS guarantees for all packets, the system applies QoS mechanisms only to high priority packets that require reliable delivery. Low priority packets are handled with best-effort service, eliminating the need for complex resource reservation and guarantee mechanisms for the entire traffic stream. This partial application of QoS reduces router complexity while maintaining service quality for critical traffic.
3Reliability
If packet prioritization mechanisms are implemented, then service quality for critical packets is improved, but processing overhead and device complexity increase
Solution Approach 1:
Packets are classified and assigned to priority queues at the point of ingress into the router, before any switching or forwarding decisions are made. This preliminary classification ensures that priority information is available for all subsequent processing stages without requiring repeated analysis. The simple queue selection logic at each processing stage reduces overhead compared to complex per-packet re-evaluation mechanisms.
Data Source
AI summary
A grid router for routing data over a grid sessionwise, involves multiple devices forwarding data according to sessions over an unreliable network between the devices. The devices include a first group of mailbox units to handle the sessions and a second group of egress units to forward session data to devices external to the network. The mailbox units and egress units are connected in a CLOS network such that each mailbox unit is accessible by each egress unit. The router further includes a distributed entity, which assigning specific sessions to specific mailbox units and to specific egress units and further defomes quality of service (QoS) levels for each session in the presence of congestion in either the network or the devices.