Context Cluster Migration for Network Device Load Balancing

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current network devices face low processing efficiency during user context migration due to processing user contexts one by one, which affects service performance during scale out or scale in operations in cloud environments.

Innovation Solution

Implementing a terminal management method that uses context clusters as units for migration, where each cluster includes multiple user contexts, allowing for simultaneous migration and recording of context locations, thereby improving migration speed and processing efficiency.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If user contexts are processed one by one during migration, then the migration process is simple to implement, but the processing efficiency is low

Engineering Contradiction:
Improveuser context migration speedVSAvoidmigration management complexity
Core Design Contradiction:
ProductivityVSDevice complexity

Solution Approach 1:

The patent merges multiple user contexts into a single context cluster that can be migrated together as one unit. Instead of migrating individual user contexts sequentially, the system groups them into clusters and migrates the entire cluster simultaneously, thereby improving migration speed while maintaining manageable complexity through standardized cluster operations.

Inventive Principle:
Principle #5Merging (Combining)

Solution Approach 2:

The patent segments the large set of user contexts into smaller manageable clusters. Each cluster contains multiple user contexts that are migrated together. This segmentation allows the system to handle migrations in controlled units rather than as a single large operation or many individual operations, balancing efficiency and complexity.

Inventive Principle:
Principle #1Segmentation

2Speed

If context clusters are used as migration units, then migration speed is improved, but the data structure becomes more complex

Engineering Contradiction:
Improvecontext migration speedVSAvoiddata structure complexity
Core Design Contradiction:
SpeedVSDevice complexity

Solution Approach 1:

The system performs preliminary actions by pre-grouping user contexts into clusters before migration occurs. The context cluster structure is established in advance, with predefined relationships between user contexts and their corresponding service processing units. This preliminary organization enables faster migration execution while the complexity is managed through standardized cluster templates and pre-computed mappings.

Inventive Principle:
Principle #10Preliminary action

Data Source

PatentEP3531730B1Terminal management method and network device
Publication Date: 2021.08.11 HUAWEI TECH CO LTD
  • EP3531730B1 patent drawingFigure 1~2
  • EP3531730B1 patent drawingFigure 3~4
  • EP3531730B1 patent drawingFigure 5~6

AI summary

The present invention provides a terminal management method and a network device. A terminal management method, wherein the method is executed by at least one service processing unit, a data dispatch unit and a control unit; the at least one service processing unit comprises a first service processing unit and a second service processing unit; each service processing unit comprises at least one context cluster, and each context cluster comprises at least one user context; and the method comprises: sending, by a control unit, a context cluster allocation message to a newly added service processing unit, wherein the context cluster allocation message comprises: a cluster identifier corresponding to a newly allocated context cluster, and a cluster allocation indication; and establishing, by the service processing unit, the context cluster in the service processing unit according to the cluster allocation indication; sending, by the control unit, a context cluster adding message to a data dispatch unit, wherein the context cluster adding message comprises: the cluster identifier corresponding to the newly allocated context cluster, a cluster adding indication, and a unit identifier of the service processing unit; and recording, by the data dispatch unit, a correspondence between the cluster identifier and the service processing unit according to the cluster adding indication, so as to forward service data corresponding to the context cluster to the service processing unit according to the correspondence.