Switch Port Congestion Management via Dynamic Rate Control
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Network congestion occurs due to high rates of write and read operations in computer networks, particularly when switches are unable to receive frames, leading to inefficiencies in data transmission.
Innovation Solution
A method and system that monitor the average transmit wait count value at switch ports to detect congestion, generating messages to notify a management module, which then reduces the rate of write and read operations to alleviate congestion by adjusting the operations rates of initiators connected to the storage system.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If the rate of write and read operations is increased to improve data access speed, then productivity is improved, but network congestion increases causing transmission inefficiency
Solution Approach 1:
The system implements feedback by monitoring transmit wait counts at switch ports and using this information to dynamically adjust operation rates. The management module receives congestion information from switches and sends control messages back to initiators to modify their read/write rates, creating a closed-loop control system that balances speed and efficiency.
Solution Approach 2:
The system dynamically adjusts the rate of read and write operations based on real-time network conditions. Instead of using a fixed operation rate, the initiators modify their rates in response to changing congestion levels, allowing the system to adapt to varying workload conditions and maintain optimal performance.
2Reliability
If continuous monitoring and control mechanisms are implemented to reduce congestion, then transmission efficiency is improved, but device complexity increases
Solution Approach 1:
The switch automatically monitors its own transmit wait counts and generates congestion control messages without requiring external intervention. The management module on the initiator side autonomously processes these messages and adjusts operation rates based on received congestion information, enabling the system to self-regulate without complex external control mechanisms.
Solution Approach 2:
The system changes operational parameters (read/write rates) based on monitored conditions rather than implementing complex structural changes. By adjusting the rate parameters dynamically, the system achieves congestion control through parameter modification rather than through complicated architectural additions.
Data Source
AI summary
In a network having a computing system interfacing with a storage system via a switch, an average transmit wait count value is compared with a first threshold value. The average transmit wait count value is updated when a switch port has a frame available for transmission but is unable to transmit the frame if a port of the storage system is unable to receive the frame or if another switch port is unable to receive the frame. A congestion message is generated for notifying a management module of congestion at the switch port when the average transmit wait count value reaches the first threshold value. Depending upon the characteristics of the congestion, a notification is sent to the computing system to either only reduce a rate of write operations to reduce congestion, or to reduce both the rate of write operations and a rate of read operations.