Bimodal Bandwidth Scheduler for Network Switch Traffic
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Network devices like switches and routers face challenges in managing bandwidth efficiently, particularly in oversubscribed conditions where recirculation traffic competes with external-link traffic, leading to potential drops in recirculation traffic throughput due to insufficient bandwidth allocation and prioritization.
Innovation Solution
Implementing a bimodal scheduling system that allocates a minimum bandwidth for recirculation traffic and prioritizes external-link traffic initially, while ensuring a guaranteed minimum rate and latency for recirculation traffic, allowing surplus bandwidth to be opportunistically used by recirculation traffic when external-link demands are met.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If recirculation traffic is serviced at lower priority than external-link traffic, then external-link bandwidth demands can be met, but recirculation traffic throughput drops after full switch bandwidth is consumed
Solution Approach 1:
The scheduler dynamically adjusts operational modes based on bandwidth availability. In a first operational mode, external-link traffic is prioritized when bandwidth is available. In a second operational mode, recirculation traffic is prioritized when external-link bandwidth demands are met, allowing the system to adapt priorities based on real-time conditions and maintain both external-link and recirculation traffic throughput.
Solution Approach 2:
The system changes the service rate parameter for recirculation traffic based on bandwidth availability. When in the first operational mode, recirculation traffic receives best-effort service at a lower service rate. When transitioning to the second operational mode, the service rate for recirculation traffic increases to ensure minimum service rate requirements are met, while external-link traffic may be throttled.
2Adaptability or versatility
If recirculation ports are treated like external ports with pooled bandwidth, then external connections can use excess bandwidth, but large memory buffers are required to accommodate traffic bursts
Solution Approach 1:
The system segments traffic into two distinct types (external-link and recirculation) with separate scheduling queues and service rates. This segmentation allows independent management of each traffic type, enabling external-link traffic to use surplus bandwidth without requiring large buffers for recirculation traffic bursts, as recirculation traffic is serviced at a guaranteed minimum rate independent of external-link demands.
3Reliability
If minimum bandwidth is allocated for recirculation traffic, then recirculation throughput is stabilized, but external-link traffic may experience delays during peak conditions
Solution Approach 1:
The scheduler transitions between operational modes based on real-time bandwidth availability and traffic conditions. When external-link bandwidth demands are low, the system operates in the second mode to guarantee recirculation traffic service. When external-link traffic is abundant, the system switches to the first mode to prioritize external-link traffic, dynamically balancing the trade-off between recirculation stability and external-link latency.
Data Source
AI summary
A system for managing bandwidth use in a device. In a specific embodiment, the device is a network device that includes a first data scheduler that is adapted to initially share available device bandwidth among a first type of traffic and a second type of traffic on an as-needed basis. A traffic monitor communicates with the first scheduler and causes the first data scheduler to guarantee predetermined transmission characteristics for the second type of traffic. The first data scheduler includes one or more routines for prioritizing first type of traffic above the second type of traffic when the network device is in a first operational mode, and prioritizing the second type of traffic above the first type of traffic when the network device is in a second operation al mode. The minimum transmission characteristics include a minimum service rate and a minimum latency for the second type of traffic.