Network Processor External Memory Protection Against Bandwidth Collapse
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Network processors face bandwidth collapse due to external memory bandwidth being outpaced by line bandwidth, leading to congestion and throughput limitations when rogue source devices or microbursts overload the system, causing all additional traffic to be dropped.
Innovation Solution
Implementing a bandwidth utilization monitoring mechanism to detect impending congestion, sending notifications to source devices to slow traffic, and automatically reducing queue sizes in unresponsive devices to prevent external memory overload, using a combination of congestion notification profiles and queue resizing strategies.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If external memory bandwidth is reduced to lower costs, then device cost decreases, but network throughput collapses when congestion occurs
Solution Approach 1:
The system performs preliminary actions by sending congestion notifications to source devices before external memory becomes fully overloaded. This proactive approach prevents bandwidth collapse by urging source devices to slow down traffic early, avoiding the need for high external memory bandwidth while maintaining network throughput during normal operation
Solution Approach 2:
The system implements feedback mechanisms by monitoring external memory bandwidth utilization and dynamically adjusting queue sizes based on congestion levels. When congestion is detected, the system reduces queue sizes to prevent overload, creating a closed-loop control system that maintains throughput without requiring excessive external memory bandwidth
2Quantity of substance
If queue sizes are increased to handle congestion, then buffer capacity increases, but external memory becomes overloaded and bandwidth collapse occurs
Solution Approach 1:
The system applies dynamics by making queue sizes adjustable rather than fixed. Queue sizes are dynamically reduced when congestion is detected and external memory utilization becomes high, preventing overload while maintaining adequate buffer capacity during normal operation. This dynamic adjustment resolves the contradiction between needing large buffers and avoiding memory overload
3Ease of operation
If traditional congestion notifications are sent to source devices, then traffic flow can be regulated, but unresponsive devices continue to overload external memory
Solution Approach 1:
The system introduces an intermediary mechanism by acting as a mediator between source devices and external memory. When source devices are unresponsive to congestion notifications, the network processor intervenes by automatically reducing queue sizes, thereby protecting external memory from overload without requiring direct cooperation from unresponsive source devices
Data Source
AI summary
Systems and methods for protecting external memory resources to prevent bandwidth collapse in a network processor. One embodiment is a network processor including an input port configured to receive packets from a source device, on-chip memory configured to store packets in queues, and external memory configured to provide a backing store to the on-chip memory. The network processor also includes a processor configured, in response to determining that the source device is unresponsive to a congestion notification, to reduce a size of one or more queues to prevent packets transferring from the on-chip memory to the external memory.