Dynamic Router Buffer Sizing for IP Network Latency
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for setting buffer sizes in IP networks are inadequate for real-time applications like VoIP and video telephony, as they are based on TCP protocols and do not account for the unique latency and packet loss requirements of UDP-based interactive applications, leading to suboptimal performance and increased packet delays.
Innovation Solution
A dynamic buffer sizing method that adjusts buffer capacities based on packet arrival rates, mean arrival rate, mean waiting time, and link speed, allowing for optimal buffer size determination without requiring outside signaling for RTT estimation, and is applicable to both real-time and non-real-time applications.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If buffer sizes are increased to minimize packet loss for TCP applications, then packet loss is reduced, but packet delay increases dramatically
Solution Approach 1:
The patent implements dynamic buffer size adjustment where the buffer size is not fixed but varies based on current network conditions. The buffer size is adjusted according to measured packet arrival rates, mean waiting times, and link speeds, allowing the system to adapt to changing traffic patterns and application requirements in real-time
Solution Approach 2:
The patent changes the buffer size parameter dynamically based on multiple factors including packet arrival rates, mean waiting times, and link speeds. By adjusting this critical parameter according to actual network conditions rather than using static values, the system optimizes the trade-off between packet loss and delay for different applications
2Loss of time
If buffer sizes are decreased to meet packet delay requirements for real-time applications, then packet delay is reduced, but packet loss increases
Solution Approach 1:
The system dynamically adjusts buffer sizes based on current network conditions including packet arrival rates and mean waiting times. This allows real-time applications to receive smaller buffers for lower delay while the system can allocate larger buffers when needed to prevent packet loss, optimizing both requirements simultaneously
Solution Approach 2:
The patent adjusts the buffer size parameter based on measured network parameters such as packet arrival rates and link speeds. By continuously monitoring and adapting this parameter, the system can meet the strict delay requirements of real-time applications like VoIP while maintaining adequate packet loss protection for TCP traffic
3Reliability
If traditional TCP-based buffer sizing methods are used for UDP real-time applications, then packet loss is minimized, but packet delay requirements are not met
Solution Approach 1:
The patent creates a universal buffer sizing method that works for both TCP and UDP applications, as well as for both real-time and non-real-time traffic. The system uses a unified approach based on packet arrival rates, mean waiting times, and link speeds that adapts to different application types without requiring separate optimization mechanisms
Solution Approach 2:
The system changes buffer size parameters based on actual measured network conditions rather than assuming TCP behavior. By using real-time measurements of packet arrival rates and waiting times, the system can optimize buffer sizes for UDP real-time applications while still providing appropriate service to TCP traffic
4Device complexity
If static buffer sizes are used in IP networks, then device complexity is reduced, but adaptability to different applications and network conditions deteriorates
Solution Approach 1:
The patent implements feedback mechanisms where the system continuously monitors network conditions including packet arrival rates, waiting times, and buffer utilization. This feedback information is used to dynamically adjust buffer sizes, allowing the system to adapt to different applications and network conditions automatically without complex manual configuration
Solution Approach 2:
The system performs self-adjustment of buffer sizes based on its own measurements of network conditions. Rather than requiring external control or complex configuration, the system uses its own operational data (packet arrival rates, waiting times) to autonomously optimize buffer parameters for different applications
Data Source
AI summary
A communications system provides a dynamic setting of optimal buffer sizes in IP networks. A method for dynamically adjusting buffer capacities of a router may include steps of monitoring a number of incoming packets to the router, determining a packet arrival rate, and determining the buffer capacities based at least partially on the packet arrival rate. Router buffers are controlled to exhibit the determined buffer capacities, e.g. during writing packets into and reading packets from each of the buffers as part of a packet routing performed by the router. In the disclosed examples, buffer size may be based on the mean arrival rate and one or more of mean packet size and mean waiting time.