Ingress Credit Scheduler for Network Bandwidth Allocation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Network devices face challenges in managing port bandwidth, particularly during oversubscription scenarios where multiple ingress ports send data packets to the same egress port, leading to bandwidth oversubscription and potential delays in packet transfer.
Innovation Solution
The method involves determining weights for each network chip based on their ingress bandwidths and using egress credit schedulers to distribute available egress port bandwidth fairly among competing ingress ports, ensuring a 'fair' distribution of bandwidth during oversubscription phases.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If multiple ingress ports send data packets to the same egress port without bandwidth management, then the network device can handle high traffic volume, but bandwidth oversubscription occurs leading to packet delays and potential buffer overflow
Solution Approach 1:
The system implements a feedback mechanism where the network device monitors the actual bandwidth usage of each ingress port and dynamically adjusts credit allocation. The egress credit scheduler receives feedback about buffer status and bandwidth consumption, then modifies credit distribution to maintain fair bandwidth allocation and prevent oversubscription-induced packet loss
Solution Approach 2:
The patent introduces credits as an intermediary mechanism to manage bandwidth allocation. Instead of directly controlling packet flow, the system uses credits as a mediator that ingress ports must acquire before transmitting packets. This credit-based intermediary layer enables fair bandwidth distribution and prevents any single ingress port from overwhelming the egress port, thereby maintaining packet transfer reliability under high traffic conditions
2Ease of operation
If bandwidth is allocated equally among all ingress ports, then fairness is maintained, but network chips with different ingress bandwidth capabilities experience unequal actual bandwidth constraints
Solution Approach 1:
The system applies local quality by configuring different credit values for each ingress port based on its specific bandwidth capability. Instead of uniform credit allocation, each network chip and ingress port receives credits proportional to its ingress bandwidth capacity. This localized differentiation ensures that ports with higher capabilities can transmit more packets while maintaining overall fairness and optimizing network throughput
3Productivity
If the egress port bandwidth is fully utilized, then maximum throughput is achieved, but any additional packets from multiple ingress ports cause buffer overflow and packet dropping
Solution Approach 1:
The system performs preliminary action by pre-allocating credits to ingress ports before packet transmission occurs. The egress credit scheduler calculates and distributes credits in advance based on configured bandwidth allocations and current network conditions. This proactive credit allocation prevents buffer overflow by ensuring that the total rate of packet arrival from all ingress ports does not exceed the egress port's processing capacity, thereby eliminating packet loss while maintaining maximum throughput
Data Source
AI summary
A method for managing port bandwidth in network devices. The method includes determining a first and a second ingress bandwidth of a first and a second network chip, respectively, determining an egress bandwidth of an egress port of a third network chip, determining a first and a second weight for the first and the second network chip, respectively, where the first and the second weight are determined based on a bandwidth including the first and second ingress bandwidth, processing a first data packet, received by a first ingress port administrated by the first network chip, based on the first weight and the egress bandwidth, and processing a second data packet, received by a second ingress port administrated by the second network chip, based on the second weight, and the egress bandwidth, where the destination of the first and the second data packet is the egress port.