Bandwidth Control Module Using Dynamic Fractional Refill Rates
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional bandwidth control methods using the leaky bucket model often result in bandwidth instability and inaccuracies due to integer constraints on refill rates, leading to bursty data transmission and deviations from intended bandwidth targets.
Innovation Solution
A bandwidth control module and method that utilize a counter and controller to cyclically adjust refill data amounts over time intervals, employing bit reversal to distribute data evenly across segments, allowing for dynamic adjustment of refill amounts to achieve more stable and accurate bandwidth control.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If integer constraints are applied to refill rates in conventional bandwidth control methods, then device complexity is reduced and ease of operation is improved, but manufacturing precision (bandwidth accuracy) deteriorates and reliability worsens due to burstiness
Solution Approach 1:
The patent applies dynamics by transitioning from static integer-based refill rates to dynamic fractional refill rates. The bandwidth control module calculates refill rates as fractions (B/C) where B and C are integers, but the actual refill amount can be fractional within each time interval. This dynamic approach allows the system to achieve precise target bandwidth values while maintaining operational simplicity through automated fractional calculation and distribution.
Solution Approach 2:
The patent changes the parameter representation from pure integers to fractions. Instead of using integer refill rates that cause bandwidth inaccuracies, the system uses fractional refill rates (B/C) where the numerator and denominator are integers but their ratio provides precise control. This parameter change enables accurate bandwidth control while maintaining the simplicity of integer-based time interval selection.
2Productivity
If large numbers of tokens are refilled in each time interval to meet target bandwidth, then productivity (bandwidth throughput) is improved, but stability deteriorates due to bursty transmission patterns
Solution Approach 1:
The patent applies segmentation by dividing the total refill amount into smaller incremental refills distributed throughout each time interval. Instead of adding all B tokens at once causing bursts, the system distributes them as smaller amounts at multiple points within the time interval C. This segmentation smooths the transmission pattern, maintaining productivity while improving bandwidth stability and reducing burstiness.
Solution Approach 2:
The patent uses periodic action by distributing refills at regular intervals within each time interval. The fractional refill mechanism creates a periodic pattern of smaller data additions rather than single large bursts. This periodic distribution maintains the required throughput while stabilizing the bandwidth by preventing concentrated burst transmissions.
3Manufacturing precision
If fractional refill rates are implemented to improve bandwidth accuracy, then manufacturing precision is improved, but device complexity increases due to additional control mechanisms
Solution Approach 1:
The patent applies self-service by implementing automated fractional calculation and distribution within the bandwidth control module. The system automatically calculates the fractional refill amount (B/C) and distributes it appropriately without requiring external intervention or complex external control mechanisms. This self-service approach achieves precise bandwidth control while limiting complexity growth through automated internal processing.
Solution Approach 2:
The patent substitutes mechanical integer-based refill mechanisms with a computational fractional refill system. Instead of using physical or mechanical means to handle integer token additions, the system uses computational calculations to determine and apply fractional refill amounts. This substitution achieves higher precision while managing complexity through software-based fractional arithmetic rather than complex hardware mechanisms.
Data Source
AI summary
A bandwidth control method is adapted for use in a network device having a system clock. The network device has a register for storing a transmittable data amount to control bandwidth. The method includes: calculating a number of elapsed periods of the system clock so as to change a counting value every predetermined time interval, the counting value being cyclic within a specific range; adjusting the transmittable data amount in the register by a first unit amount when the counting value reaches a first count value; and adjusting the transmittable data amount in the register by a second unit amount after adjusting the transmittable data amount by the first unit amount and elapse of the predetermined time interval and when the counting value corresponds to a second count value. The first count value is different from the second count value, and the first unit amount is different from the second unit amount.