Slow eNodeB Identification and Queue Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In telecom networks, slow eNodeB/HNBs cause congestion, delays, and packet drops due to non-ideal backhaul links, making it impractical to reserve resources for each association, especially when thousands of connections are terminated by a core network node.
Innovation Solution
A method to identify slow eNodeBs/HNBs by detecting full message queues, high round-trip times, and low throughput, and then moving them into a dedicated slow queue to mitigate the impact, allowing non-slow members to progress at their regular pace and enabling applications to throttle signaling and downlink data accordingly.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If resources are reserved for each SCTP association, then service quality for each association is improved, but device complexity and resource consumption increase significantly
Solution Approach 1:
The patent segments the SCTP associations into two distinct groups: slow associations and fast associations. This segmentation allows the core network node to apply different resource allocation strategies to each group, reserving dedicated resources for slow associations while sharing resources more aggressively for fast associations, thereby reducing overall resource consumption while maintaining service quality for affected users.
Solution Approach 2:
The patent extracts slow associations from the general pool of SCTP associations and places them into a separate dedicated queue. By taking out the problematic slow associations, the system isolates their resource consumption patterns, allowing other fast associations to access shared resources more efficiently without being constrained by the resource reservation requirements of slow associations.
2Reliability
If a dedicated queue is created for slow eNodeBs, then impact mitigation is improved, but device complexity increases
Solution Approach 1:
The patent implements segmentation by creating a separate slow association queue distinct from the existing fast association queue. This structural division allows the core network node to independently manage slow associations, applying specific mitigation strategies such as reduced resource allocation or priority adjustments without affecting the operation of fast associations, thereby containing complexity growth to a localized scope.
Solution Approach 2:
The slow association queue acts as an intermediary structure between the core network node and slow eNodeBs. This intermediary queue absorbs the complexity of managing slow associations by providing a dedicated buffer and processing path, isolating the complexity from the main network processing logic and enabling simpler, more scalable management of slow connections.
Data Source
AI summary
A method and computer readable medium for identifying slow base stations and providing impact mitigation are described. In one embodiment, the method includes detecting that a first base station, using a first queue, is slow, wherein a slow base station is a base station that that cannot keep up with the rate at which a core node is generating update messages over a prolonged period; providing a slow base station queue; and moving the first base station from the first queue to the slow base station queue.