Switch Meter Conformance via Dynamic Sampling
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Traditional methods for monitoring bandwidth usage and determining conformance with classifier policies in switches are inaccurate, slow to reflect changes, and require significant computational overhead due to variations in rate-limiting parameters and sample periods.
Innovation Solution
A system that determines a sampling interval based on token allocation frequencies and configuration parameters, calculating performance rates such as conform, exceed, and violation rates by querying the switch hardware at predetermined intervals, thereby adapting to short or long bursts of traffic with low computational overhead.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional methods are used to monitor bandwidth usage and determine conformance with classifier policies, then comprehensive monitoring coverage is achieved, but measurement precision and response speed deteriorate due to inaccuracies and slowness in reflecting changes
Solution Approach 1:
The patent implements periodic sampling of traffic data at dynamically determined intervals. The sampling interval is calculated based on token allocation frequencies and configuration parameters, creating a periodic measurement mechanism that balances accuracy with response time. This periodic action allows the system to capture traffic changes at appropriate moments without continuous monitoring overhead.
Solution Approach 2:
The sampling interval is made dynamic rather than fixed, adapting to different token allocation frequencies and traffic conditions. The system calculates optimal sampling intervals based on configuration parameters, allowing the measurement precision and response speed to be optimized for different traffic scenarios and meter configurations.
2Reliability
If traditional monitoring methods are used with variations in rate-limiting parameters and sample periods, then comprehensive traffic analysis is achieved, but device complexity and computational overhead increase significantly
Solution Approach 1:
The patent changes the sampling interval parameter dynamically based on token allocation frequencies and configuration parameters. By adjusting this key parameter, the system maintains reliable traffic conformance determination while reducing computational overhead. The parameter change allows the system to adapt to different traffic conditions without increasing overall complexity.
Solution Approach 2:
The patent extracts only the necessary sampling operations at optimally determined intervals, rather than continuously monitoring all traffic parameters. This extraction of essential measurements at strategic moments reduces computational overhead while maintaining the reliability needed for accurate traffic conformance determination.
3Speed
If frequent sampling is performed to capture traffic changes quickly, then response speed improves, but use of energy and computational resources increases
Solution Approach 1:
The system performs sampling at periodic intervals determined by token allocation frequencies rather than continuously or at fixed high frequencies. This periodic action captures traffic changes at appropriate moments while significantly reducing the energy and computational resource utilization compared to continuous monitoring.
Solution Approach 2:
The system performs sampling at the minimum necessary frequency to capture traffic changes effectively, rather than excessive frequent sampling. The sampling interval is calculated to be sufficient for detecting relevant traffic patterns while minimizing resource consumption, applying partial action only where needed.
Data Source
AI summary
A system for facilitating data transmission in a switch is provided. During operation, the system can obtain one or more sets of configuration parameters for a plurality of traffic policing filters of the switch. A respective traffic policing filter can correspond to a token bucket. A number of tokens in the token bucket can indicate whether to forward a packet associated with the traffic policing filter. The system can determine the token allocation frequencies for the plurality of traffic policing filters based on the one or more sets of configuration parameters. The system can then select a sampling interval from the token allocation frequencies based on a selection policy and determine a performance rate for the plurality of traffic policing filters based on the sampling interval.