Random Early Discard Mechanism for Network Load Management
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing network devices face inefficiencies and instabilities due to the juxtaposition of high-speed and low-speed packets, leading to potential overloads that can stop data transfer, as they lack explicit supervisory functions over data transport rates, and struggle with unpredictable network traffic loads exceeding processing capabilities.
Innovation Solution
Implementing an enhanced random early discard mechanism in network devices that intelligently biases packet drop probabilities based on application-aware and flow-aware metrics or state conditions to manage network loads more effectively.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If random early discard is implemented to manage network load, then network stability is improved, but packet loss increases
Solution Approach 1:
The patent applies different discard probabilities to different packets based on their classification, source, destination, or flow characteristics. Instead of uniform random discard, the system selectively discards packets with higher probability based on local attributes, thereby maintaining network stability while minimizing impact on critical traffic.
Solution Approach 2:
The system dynamically adjusts the discard probability parameter based on network conditions, packet classification, and flow state. By changing the probability parameter selectively for different packets rather than applying a fixed uniform probability, the system achieves load management with reduced overall packet loss.
2Productivity
If explicit data rate control is implemented to optimize bandwidth allocation, then bandwidth utilization is improved, but device complexity increases
Solution Approach 1:
The patent segments network traffic into different classes or flows based on classification criteria, and applies different rate control policies to each segment. This segmentation allows efficient bandwidth allocation without requiring complex centralized control for all traffic, as each segment can be managed independently with simpler rules.
Solution Approach 2:
The system performs preliminary classification and marking of packets before they enter the queue, so that subsequent rate control and discard decisions can be made using simple pre-defined rules rather than complex real-time analysis. This preliminary action reduces the computational complexity during packet processing.
Data Source
AI summary
Methods, apparatuses and systems directed to enhanced random early discard mechanisms implemented in various networked devices including end-systems such as servers and intermediate systems such as gateways and routers. In one implementation, the present invention enables a random early discard mechanism that intelligently biases the drop probabilities of select packets based on one or more application-aware and/or flow-aware metrics or state conditions.