Intermediary Device Path Selection for Bursty Packet Drops
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Bursty packet drops in networks significantly degrade the performance of applications like VOIP, video telephony, and online gaming, as existing solutions like forward error correction and retransmit buffers fail to provide adequate feedback, leading to unawareness of packet loss and subsequent network overcapacity and quality degradation.
Innovation Solution
Intermediary devices monitor packet loss and classify applications based on sensitivity, selecting communication paths with appropriate threshold path quality and retransmitting a subset of packets to minimize bursty packet drop effects, hiding burstiness from clients and servers by distributing packet losses randomly.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If forward error correction and retransmit buffers are used to address packet loss, then packet loss recovery is improved, but network overcapacity and quality degradation occur due to lack of feedback
Solution Approach 1:
The patent introduces a feedback mechanism where intermediary devices monitor packet loss and provide information back to clients and servers. This enables dynamic adjustment of transmission rates and buffer management, preventing network overcapacity while maintaining reliable packet loss recovery through informed retransmission decisions.
Solution Approach 2:
The patent employs intermediary devices positioned between clients and servers that act as mediators. These intermediaries monitor network conditions, manage retransmit buffers, and control packet retransmission, thereby improving packet loss recovery while preventing network overcapacity by regulating traffic flow through the intermediary layer.
2Reliability
If packets are retransmitted without feedback to clients and servers, then application performance is improved, but network overcapacity occurs
Solution Approach 1:
The patent implements feedback loops where intermediary devices monitor the effectiveness of retransmissions and adjust future retransmission decisions accordingly. This feedback mechanism ensures that retransmissions improve application performance without causing network overcapacity by learning from past transmission outcomes and adapting transmission rates.
Solution Approach 2:
The intermediary devices autonomously manage packet retransmission decisions based on monitored network conditions and application performance metrics. This self-service capability allows the system to optimize retransmission behavior without external intervention, improving application performance while preventing network overcapacity through autonomous rate control.
3Productivity
If communication paths are selected based on highest performance metrics, then throughput is improved, but bursty packet drops severely degrade application performance
Solution Approach 1:
The patent applies different quality thresholds to different communication paths based on their characteristics and the sensitivity of applications using them. Rather than uniformly selecting the highest throughput path, the system evaluates path quality locally for each application type, selecting paths that meet minimum quality thresholds appropriate for burst-sensitive applications while maintaining overall network throughput.
Solution Approach 2:
The patent dynamically changes the selection criteria for communication paths based on application sensitivity to bursty packet drops. For burst-sensitive applications, the system modifies path selection parameters to prioritize stability and consistency over maximum throughput, while for less sensitive applications, it returns to throughput-optimized path selection.
Data Source
AI summary
Described embodiments provide systems and methods for selecting communication paths for applications sensitive to bursty packet drops. A device intermediary to a client and a server may identify an application for which packets are to be communicated between the client and the server. The device may determine a sensitivity level of the application to a network disruption affecting the packets. The device may estimate, for each path between the client and the server for communicating the one or more packets, a path quality for the path indicating a likelihood that the network disruption affects the one or more packets. The device may select path for communicating the packets based on the sensitivity level of the application and the path quality. The device may communicate the packets between the client and the server via the path.