Decentralized Load Balancing in Network Management Clusters

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current network management systems face challenges in load balancing due to stateful session semantics, requiring manual or centralized algorithms for redistributing network elements, which are unwieldy and difficult to automate, especially in large and heterogeneous mobile communication networks.

Innovation Solution

A method and apparatus for decentralized dynamic load balancing across network management processing elements, where a cluster head monitors and rebalances the load by updating the cluster state, suspending operations, and reallocating tasks among members once tasks are completed, using a load balancing algorithm to distribute processing load based on shared data.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Extent of automation

If manual or centralized algorithms are used for redistributing network elements, then load balancing can be achieved, but the system becomes unwieldy and difficult to automate

Engineering Contradiction:
Improveautomation of load balancingVSAvoidcomplexity of redistribution system
Core Design Contradiction:
Extent of automationVSDevice complexity

Solution Approach 1:

The system enables autonomous load balancing where cluster members automatically detect load imbalances and initiate redistribution without centralized control. Each member monitors its own load and autonomously decides when to suspend operations and transfer network elements, making the system self-regulating and eliminating the need for complex centralized management algorithms

Inventive Principle:
Principle #25Self-service

Solution Approach 2:

The load balancing functionality is segmented into independent cluster members that operate autonomously. Rather than a single centralized controller, the system divides the management function across multiple distributed nodes, each capable of independent decision-making. This segmentation reduces the complexity of any single component while maintaining overall system coordination

Inventive Principle:
Principle #1Segmentation

2Productivity

If cluster head updates cluster state to initiate rebalancing, then load distribution is optimized, but processing interruptions occur during state transitions

Engineering Contradiction:
Improveload balancing efficiencyVSAvoidprocessing interruption time
Core Design Contradiction:
ProductivityVSLoss of time

Solution Approach 1:

The cluster head proactively monitors load distribution and updates the cluster state before significant imbalances occur. By detecting load conditions in advance and initiating rebalancing operations proactively, the system prevents extreme load variations that would require more extensive redistribution, thereby reducing the duration of processing interruptions

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The system implements dynamic load balancing where the cluster state and member operations adapt in real-time based on current load conditions. Members can dynamically suspend and resume operations based on cluster state updates, allowing the system to flexibly respond to changing load patterns while minimizing overall disruption to processing

Inventive Principle:
Principle #15Dynamics

Data Source

PatentEP3061207B1Load balancing in a distributed network management architecture
Publication Date: 2018.08.08 TELEFONAKTIEBOLAGET LM ERICSSON (PUBL)
  • EP3061207B1 patent drawingFigure 1
  • EP3061207B1 patent drawingFigure 2
  • EP3061207B1 patent drawingFigure 3~4

AI summary

A method of distributing network management processing load across a plurality of network management processing elements is disclosed. Each network management processing element is a member of a cluster, one member being a head of the cluster updating the cluster state,and members of the cluster following the cluster state. The method comprises the cluster head monitoring the network management processing load across the members of the cluster. The method further comprises, upon detecting that the cluster load is unbalanced, the cluster head updating the cluster state to initiate automatic rebalancing of the network management processing load across at least a subset of the plurality of members of the cluster,once tasks being processed by the subset of the plurality of members have been completed. Also disclosed are a method of distributing the processing for controlling a communication network, and a network management processing element.