Switching Fabric Bandwidth Allocation and Advertisement
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Network devices continue to advertise full bandwidth capacity despite switching fabric degradation, leading to over-subscription and dropped network traffic due to failure to account for reduced bandwidth caused by hardware, software errors, or environmental conditions.
Innovation Solution
Implement a system that detects switching fabric degradation, allocates reduced bandwidth among interfaces, determines unreserved bandwidth, and advertises this information to neighbor network devices using Interior Gateway Protocol (IGP) messages to ensure accurate bandwidth reporting and prevent traffic loss.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If network devices continue to advertise full bandwidth capacity, then bandwidth utilization appears maximized, but traffic loss occurs due to over-subscription when switching fabric degradation is not accounted for
Solution Approach 1:
The system implements feedback by continuously monitoring switching fabric performance metrics (error rates, packet loss, latency) and using this information to dynamically adjust bandwidth advertisements. When degradation is detected, the network device reduces the advertised bandwidth capacity to match actual available capacity, preventing over-subscription while maintaining accurate bandwidth utilization reporting.
Solution Approach 2:
The patent applies dynamics by making bandwidth advertisements adaptive rather than static. The advertised bandwidth capacity changes dynamically based on real-time switching fabric conditions, allowing the system to respond to degradation events and recover from them when the fabric returns to normal operation, thus maintaining both reliability and productivity.
2Reliability
If network devices advertise reduced bandwidth capacity due to switching fabric degradation, then traffic loss is prevented, but network performance appears reduced due to conservative bandwidth reporting
Solution Approach 1:
The system dynamically adjusts bandwidth advertisements based on real-time switching fabric performance, allowing the network to operate at full capacity when healthy and automatically reduce capacity only when degradation is detected. This dynamic approach ensures reliability is maintained during degradation events while maximizing productivity during normal operation.
Solution Approach 2:
The system takes preliminary action by proactively reducing bandwidth advertisements before over-subscription can occur. By monitoring switching fabric metrics and preemptively adjusting capacity announcements, the system prevents traffic loss rather than reacting to it after the fact, thus maintaining reliability without unnecessary conservatism.
3Measurement precision
If network devices monitor and respond to switching fabric degradation, then accurate bandwidth reporting is achieved, but system complexity increases due to additional monitoring and control mechanisms
Solution Approach 1:
The system implements self-service by having the switching fabric and network device automatically monitor their own performance metrics and adjust bandwidth advertisements without external intervention. The network device uses built-in monitoring capabilities to detect degradation and autonomously modifies capacity announcements, reducing the need for complex external monitoring infrastructure while maintaining high measurement precision.
Solution Approach 2:
The patent uses feedback mechanisms where the network device monitors its own switching fabric performance and uses this internal feedback to adjust bandwidth advertisements. This self-contained feedback loop achieves accurate bandwidth reporting without requiring complex external monitoring systems, as the device uses its own operational data to make precise capacity assessments.
4Measurement precision
If bandwidth advertisements are updated frequently in response to switching fabric changes, then accurate real-time bandwidth information is provided, but network overhead increases due to frequent protocol messages
Solution Approach 1:
The system employs periodic action by updating bandwidth advertisements at intervals based on switching fabric stability rather than continuously. When the fabric performance remains stable, updates are suppressed; when degradation is detected, updates are triggered. This periodic update approach maintains accurate bandwidth information while minimizing unnecessary protocol messages and reducing network overhead.
Solution Approach 2:
The update frequency is made dynamic, adjusting based on switching fabric conditions. During stable operation, updates are infrequent to reduce overhead. When degradation events occur, the system dynamically increases update frequency to provide accurate real-time bandwidth information. This dynamic approach balances measurement precision with energy efficiency by matching update intensity to actual network conditions.
Data Source
Figure 1A
Figure 1B
Figure 1C
AI summary
A network device may detect a reduced switching fabric bandwidth due to switching fabric degradation of a switching fabric. The network device may allocate the reduced switching fabric bandwidth to one or more interfaces of a packet processor. The network device may determine a first maximum reservable bandwidth for an interface of the one or more interfaces. The network device may identify a reserved bandwidth for the interface. The network device may determine an unreserved bandwidth for the interface based on the first maximum reservable bandwidth and the reserved bandwidth. The network device may advertise the unreserved bandwidth, for the interface, to a neighbor network device that communicates with the network device via the interface. The network device may provide an instruction, to the neighbor network device, for the neighbor network device to update a second maximum reservable bandwidth associated with the neighbor network device