Dynamic Throughput Measurement for Burst Traffic
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for measuring data throughput in communication connections, particularly in burst traffic scenarios, fail to accurately distinguish between 'good' and 'bad' idle times, leading to inaccurate throughput measurements due to the inclusion of all idle times, which can significantly falsify results and do not reflect the actual user experience.
Innovation Solution
A method that determines data throughput by setting a dynamic end time based on user data transmission in the opposite direction, allowing for the differentiation between 'good' and 'bad' idle times, where 'good' idle times are discarded and 'bad' idle times are accounted for, using a connection-oriented protocol like TCP or packet-oriented protocols like UDP, with the start time defined by the transmission of the first useful data or acknowledgment messages.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If all idle times are discarded during throughput measurement, then the measurement process is simple, but the measurement precision deteriorates because good and bad idle times are not distinguished
Solution Approach 1:
The patent segments idle times into two distinct categories: good idle times (caused by send buffer emptying) and bad idle times (caused by network issues). This segmentation allows differential treatment of idle periods, improving measurement accuracy while maintaining operational simplicity by applying different rules to different segments of idle time.
Solution Approach 2:
Instead of discarding all idle times uniformly, the patent inverts the approach by selectively including only bad idle times in the measurement while excluding good idle times. This inversion of the traditional all-or-nothing approach resolves the contradiction by achieving both simplicity and precision through targeted inclusion rather than blanket exclusion.
2Ease of manufacture
If a fixed measurement interval is used, then the measurement method is simple to implement, but the measurement precision deteriorates for burst traffic due to inclusion of rest periods
Solution Approach 1:
The patent transitions from a static fixed measurement interval to a dynamic measurement interval that adapts to burst traffic patterns. The interval dynamically starts at the first data packet and ends at the last data packet, automatically adjusting to capture only active transmission periods while excluding rest periods, thereby maintaining simplicity while improving precision for burst traffic scenarios.
3Duration of action of moving object
If measurement includes all idle times, then the measurement covers the entire period, but the productivity of actual data transmission is reduced due to falsified results
Solution Approach 1:
The patent extracts and removes good idle times from the measurement period, separating them from bad idle times. By taking out the non-productive good idle periods caused by send buffer emptying, the measurement focuses only on periods that reflect actual transmission performance, thereby maintaining comprehensive coverage while improving the accuracy of productivity representation.
Data Source
Figure 1
Figure 2
AI summary
The method involves determining a number of transmitted user data in a transmission direction within a defined time interval. The time interval is determined by a fixed or dynamic start timepoint and a dynamic end timepoint. The end timepoint of the time interval is determined by transmitting the user data in another transmission direction. The start timepoint is determined by forwarding or confirmation of the user data in the former direction. The user data is selected in the former direction during the time interval based on the confirmed user data. Independent claims are also included for the following: (1) a device for performing a method for determining data throughput of a connection oriented communication connection between a client and a server of a communication system via a bidirectional transmission path in a transmission direction (2) a program code stored on a computer readable medium comprising instructions for performing a method for determining data throughput of a connection oriented communication connection between a client and a server of a communication system via a bidirectional transmission path in a transmission direction.