Dynamic Priority Queue Depth Adjustment for Network Switch Memory
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing network devices allocate fixed queue depths for priority queues, leading to unused memory when network traffic is not segmented into all priority levels, rendering the unused memory unavailable for other queues.
Innovation Solution
Implementing a queue controller that dynamically adjusts the maximum depth of priority queues based on network traffic parameters, such as priority levels, to optimize performance and balance traffic delivery needs.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If fixed queue depths are allocated at startup, then memory allocation is simple and deterministic, but unused memory space is unavailable to other queues when not all priority levels are used
Solution Approach 1:
The patent implements dynamic queue depth adjustment where the queue controller continuously monitors traffic patterns and modifies queue depths in real-time based on actual usage. This transforms the static fixed allocation into a dynamic system that adapts to changing traffic conditions, allowing memory to be reallocated from unused priority queues to actively used ones without requiring manual reconfiguration.
Solution Approach 2:
The system changes the queue depth parameter dynamically based on traffic conditions. The queue controller monitors metrics such as queue occupancy, traffic rate, and priority level utilization, then adjusts the depth parameter accordingly. This allows the same memory pool to be efficiently distributed across different priority queues based on actual demand rather than predetermined fixed allocations.
2Stability of the object's composition
If unused allocated memory space is reserved for future use, then memory allocation is stable and predictable, but the memory cannot be reallocated to currently active priority queues
Solution Approach 1:
The queue controller implements a feedback mechanism that continuously monitors queue utilization metrics and uses this information to dynamically adjust queue depths. When a priority queue is not fully utilized, the controller detects this condition and reallocates the unused memory to other active queues, thereby improving overall traffic handling efficiency while maintaining allocation stability through controlled adaptation.
Solution Approach 2:
The system performs self-service by automatically monitoring its own queue utilization and reallocating memory resources without external intervention. The queue controller autonomously detects unused memory in priority queues and redistributes it to active queues based on current traffic demands, eliminating the need for manual reconfiguration while maintaining stable and efficient operation.
3Quantity of substance
If dynamic queue depth adjustment is implemented, then memory utilization is optimized, but system complexity increases due to additional control mechanisms
Solution Approach 1:
The patent merges the queue depth control functionality into the existing queue controller that already manages packet scheduling and priority handling. By integrating the dynamic depth adjustment logic with the existing control structure, the system achieves optimized memory utilization without adding separate complex control mechanisms. The queue controller simultaneously performs scheduling decisions and memory allocation adjustments based on unified traffic state information.
Data Source
AI summary
A method of modifying a priority queue configuration of a network switch is described. The method comprises configuring a priority queue configuration, monitoring a network parameter, and adjusting the priority queue configuration based on the monitored network parameter.