Distributed Queue System for Telecommunications Network Fault Tolerance

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Telecommunications networks face challenges in maintaining high availability and avoiding single points of failure, especially with increased complexity from virtual machine migration and dynamic workloads, requiring enhanced network control and management to ensure reliability.

Innovation Solution

A fault-tolerant queue system is implemented using a distributed database with N-way redundancy, supporting multi-producer and multi-subscriber capabilities, along with distributed synchronization and locking, allowing for replayable and prioritized message handling with independent consumer groups and application-specific policies.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Ease of operation

If a centralized queue management system is used, then message processing is simplified, but single points of failure are created reducing network reliability

Engineering Contradiction:
Improvequeue management simplicityVSAvoidnetwork availability
Core Design Contradiction:
Ease of operationVSReliability

Solution Approach 1:

The queue management system is segmented into multiple independent queue databases distributed across different network nodes. Each queue database maintains a portion of the queue data and can operate independently, eliminating the single point of failure while collectively providing complete queue management functionality.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

Queue data is replicated across multiple queue databases in the distributed network. Each queue database maintains copies of the queue data, ensuring that if one database fails, the data and functionality are preserved in other databases, thereby eliminating single points of failure.

Inventive Principle:
Principle #26Copying

2Reliability

If multiple consumer replicas are deployed for fault tolerance, then network reliability is improved, but system complexity increases

Engineering Contradiction:
Improvefault toleranceVSAvoidsystem architecture complexity
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

Multiple consumer replicas are merged into a coordinated distributed system that shares state and processing logic. The replicas work together as a unified consumer group, managing their coordination through standardized protocols, which reduces the effective complexity compared to managing independent systems.

Inventive Principle:
Principle #5Merging (Combining)

Solution Approach 2:

The system uses configurable parameters such as replication factors and consumer group settings to control the degree of redundancy and coordination. By adjusting these parameters, the system can optimize between fault tolerance and complexity based on specific operational requirements.

Inventive Principle:
Principle #35Parameter changes

3Reliability

If queue data is replicated across distributed databases, then no single point of failure exists, but synchronization and locking complexity increases

Engineering Contradiction:
Improveelimination of single points of failureVSAvoidsynchronization complexity
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

A coordination service acts as an intermediary between distributed queue databases and consumer replicas. This mediator manages synchronization, locking, and state consistency across the distributed system, centralizing the complexity of coordination while keeping individual queue database implementations simple.

Inventive Principle:
Principle #24Intermediary (Mediator)

Solution Approach 2:

The distributed queue system implements feedback mechanisms where queue databases and consumer replicas continuously exchange state information and coordination signals. This feedback loop enables automatic synchronization and conflict resolution, reducing the manual complexity of maintaining consistency across distributed databases.

Inventive Principle:
Principle #23Feedback

Data Source

PatentEP2963548B1Method for enhancing the reliability of a telecommunications network, system, telecommunications network and program
Publication Date: 2022.08.31 DEUTSCHE TELEKOM AG
  • EP2963548B1 patent drawingFigure 1~2

AI summary

The invention relates to a method for enhancing the reliability of a telecommunications network, the telecommunications network comprising a plurality of network nodes running at least a first network process and a second network process, wherein a message or a request of a specific kind, being directed from the first network process to the second network process, is processed by being produced as a specific element of a queue, wherein the plurality of elements of the queue are able to be produced by producer network processes and are able to be -- read or -- read and consumed by at least a first consumer network process and a second consumer network process, wherein a first and second private read indicator are assigned to the first and second consumer network process, wherein a global read indicator is assigned to the first and second consumer network processes, wherein a first and second private consume indicator are assigned to the first and second consumer network process, wherein a global consume indicator is assigned to the first and second consumer network processes.