Bursty Traffic Detection and Multipath Scheduling
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing multipath communication systems face challenges in efficiently detecting bursty traffic patterns and scheduling data traffic to avoid unnecessary use of expensive data paths, particularly in scenarios where cheaper paths are sufficient to meet demand without degrading Quality of Experience (QoE).
Innovation Solution
A data traffic analyzer measures and analyzes data traffic parameters, such as volume and protocol data unit count, along with last path usage timestamps, to detect bursty traffic patterns by comparing them to a time interval threshold, triggering a bursty traffic signal to delay data transmission on more expensive paths, thereby optimizing the use of cheaper paths during peak periods.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If the video source uses all available data paths to fill the sink buffer as fast as possible, then the buffer filling speed is improved, but the use of expensive data paths increases unnecessarily
Solution Approach 1:
The system performs preliminary detection of bursty traffic patterns using the data traffic analyzer before committing to use expensive data paths. By analyzing traffic parameters (volume, PDU count, timestamps) and comparing them to thresholds, the system identifies opportunities to delay overflow traffic until cheaper paths can accommodate it, thus avoiding unnecessary use of expensive paths while maintaining buffer filling speed
Solution Approach 2:
The multipath scheduler receives feedback from the data traffic analyzer about bursty traffic patterns and adjusts its scheduling decisions accordingly. When bursty patterns are detected, the scheduler delays overflow traffic to be transmitted later when cheaper paths are available, creating a feedback loop that optimizes path selection dynamically
2Loss of energy
If the cheaper path is used to satisfy demand, then the cost is reduced, but the capacity utilization of the cheaper path may be insufficient during peak bursts
Solution Approach 1:
The system dynamically adjusts path selection based on real-time traffic conditions. The data traffic analyzer continuously monitors traffic parameters and the multipath scheduler dynamically switches between using cheaper paths for baseline traffic and expensive paths for peak bursts, optimizing both cost and reliability adaptation
Solution Approach 2:
The system delays overflow traffic temporarily rather than immediately using expensive paths. By buffering overflow traffic and transmitting it later when cheaper paths have capacity, the system maintains reliability while reducing cost
3Loss of energy
If bursty traffic patterns are detected and overflow traffic is delayed, then the use of expensive paths is reduced, but the detection complexity increases
Solution Approach 1:
The detection mechanism is segmented into distinct functional components: the data traffic analyzer handles parameter measurement and pattern detection, while the multipath scheduler handles scheduling decisions. This segmentation allows each component to be optimized independently and simplifies the overall system architecture
Solution Approach 2:
The data traffic analyzer operates autonomously to detect bursty patterns and provide recommendations to the multipath scheduler. The system self-monitors traffic parameters and automatically triggers scheduling adjustments without requiring complex external control mechanisms
Data Source
AI summary
A data traffic analyzer is configured for detecting bursty data traffic on a data path. The data traffic analyzer measures data traffic parameters of the data path and analyzes the data traffic parameters of the data path to detect a bursty traffic pattern based on comparing the data traffic parameters to a time interval threshold TLimit which defines gaps between bursty data traffic. The measured traffic parameters include: i) a data traffic volume and/or a protocol data unit (PDU) count transmitted on the data path; and ii) a time stamp TLPU characterizing the last path usage (LPU) of the data path.