EPC Network Failure Prevention via Load Sharing and Traffic Throttling
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The 4G EPC network is susceptible to catastrophic failures due to unconsidered failure scenarios, leading to network outages that can disconnect hundreds of thousands of user equipment (UEs) and cause message surges that overload authentication and reauthorization devices, resulting in service disruptions.
Innovation Solution
Implementing mechanisms for network nodes to share load and capacity information, throttle traffic, divert traffic to alternate nodes, and deny connection requests when the network is approaching failure, thereby preventing simultaneous reconnections and alleviating congestion.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If the EPC network uses standard 4G network nodes with reduced space and power requirements, then the network can handle millions of UEs with minimal resources, but the network becomes susceptible to catastrophic failures during message surges
Solution Approach 1:
The system performs preliminary actions by having network nodes continuously monitor their own load and capacity information, and proactively throttle or divert traffic before failure occurs. The HSS receives advance warnings from MMEs about approaching capacity limits and preemptively manages connection requests to prevent the catastrophic failure scenario where all disconnected UEs attempt to reconnect simultaneously.
2Adaptability or versatility
If the HSS handles authentication requests for millions of UEs, then the network can provide comprehensive service coverage, but the HSS becomes overloaded during message surges and cannot sustain authentication requests
Solution Approach 1:
The system implements feedback mechanisms where MMEs continuously report their load and capacity information to the HSS. The HSS uses this feedback to dynamically adjust its authentication request handling, throttling requests when approaching capacity limits and allowing them when capacity is available. This closed-loop feedback prevents the HSS from being overwhelmed during message surges while maintaining comprehensive service coverage.
3Reliability
If network nodes share load and capacity information to prevent failures, then the network can maintain service continuity, but the system complexity increases due to additional information sharing mechanisms
Solution Approach 1:
The system achieves universality by having network nodes (MMEs, HSS) perform multiple functions: they not only handle their primary authentication and routing functions but also simultaneously monitor their own load, report capacity information, and participate in collective failure prevention. This multi-functionality approach allows the network to maintain service continuity through information sharing without requiring entirely separate complex systems.
Data Source
AI summary
A first device, provided in an evolved packet core (EPC) network, receives load information associated with a second device provided in the EPC network. The load information includes a load threshold associated with the second device. The first device also receives traffic to be provided to the second device, and determines whether providing the traffic to the second device will cause the second device to exceed the load threshold. The first device provides the traffic to the second device when providing the traffic to the second device will not cause the second device to exceed the load threshold. The first device reduces a quantity associated with the traffic, to create throttled traffic, and provides the throttled traffic to the second device, when providing the traffic to the second device will cause the second device to exceed the load threshold.