Message Discard Algorithm for Network Congestion Management
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Computer network traffic congestion often leads to denial of service access due to the inability to timely process authentication messages, necessitating the discarding of less important messages while allowing important ones, but existing methods lack a nuanced approach to determine which messages to discard and which to allow.
Innovation Solution
Implementing pluggable traffic congestion policies that utilize message rates, congestion levels, and policy-defined priority values to determine whether to allow or discard messages, with algorithms that progressively limit message rates based on priority values as congestion levels increase.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If messages are discarded during congestion to reduce traffic load, then network congestion is reduced, but service accessibility deteriorates because authentication messages may be discarded
Solution Approach 1:
The patent applies local quality by assigning different discard priorities to different message types based on their importance. Critical messages (e.g., authentication, billing, roaming) are protected from discard while non-critical messages are discarded during congestion. This is implemented through message type identification and selective discard policies that treat different parts of the traffic stream differently based on their local characteristics.
Solution Approach 2:
The patent changes the parameter of message discard probability based on congestion level and message priority. During different congestion scenarios, the system dynamically adjusts which messages are discarded by changing the discard parameter selectively for different message types, rather than applying a uniform discard policy to all messages.
2Device complexity
If uniform message discard policy is applied during congestion, then implementation complexity is reduced, but message handling precision deteriorates because all messages are treated equally
Solution Approach 1:
The patent implements local quality by creating different discard policies for different message types. Instead of a uniform approach, the system identifies message types (authentication, billing, roaming, local, etc.) and applies locally optimized discard decisions based on each type's criticality, achieving precise message selection without excessive complexity.
Solution Approach 2:
The system changes the discard parameter based on message type and congestion scenario. By parameterizing the discard decision with message type as a variable, the patent achieves precise control over which messages are discarded while maintaining manageable complexity through structured policy definitions.
3Reliability
If important messages are prioritized during congestion, then service reliability is improved, but message rate limitation effectiveness deteriorates because high priority messages consume more bandwidth
Solution Approach 1:
The patent applies local quality by protecting specific critical message types (authentication, billing, roaming) from discard while allowing non-critical messages to be discarded. This localized protection ensures authentication reliability without requiring all messages to be treated as high priority, thus maintaining congestion mitigation efficiency.
Solution Approach 2:
The system dynamically changes the discard probability parameter based on message type and congestion level. Critical messages have their discard parameter set to zero or low values during congestion, while non-critical messages have higher discard probabilities, achieving both reliability for important services and overall congestion mitigation.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
The subject matter described herein relates to methods, systems, and computer readable media for discarding messages during a congestion event. One method includes registering a traffic congestion policy for handling traffic associated with an application during congestion. The method further includes determining a first congestion level associated with a congestion event. The method also includes determining message rates of messages associated with similar message priority values, wherein the message priority values are determined using the traffic congestion policy. The method further includes discarding a first message using the message rates, the first congestion level, and a message discard algorithm, wherein the message discard algorithm is determined using the traffic congestion policy.