Shared Registration Data Store for Network Node Clusters

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

IP telephony networks face scalability issues with large numbers of subscribers, as traditional sharding techniques increase routing complexity and redundancy requirements, making it difficult to manage state information and provide efficient network scalability.

Innovation Solution

Implementing a cluster of network nodes that can process messages for any subscriber, using a shared registration data store and cache to distribute registration data, allowing any node to handle requests and reducing the need for longstanding associations between subscribers and specific nodes, thereby enhancing scalability and reducing complexity.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Quantity of substance

If sharding techniques are employed to allocate subscribers to specific network nodes, then memory requirements per node are reduced, but routing complexity increases

Engineering Contradiction:
Improvememory requirements per nodeVSAvoidrouting complexity
Core Design Contradiction:
Quantity of substanceVSDevice complexity

Solution Approach 1:

The patent divides the network into clusters of network nodes, where each cluster serves a specific geographic region or subscriber group. This segmentation allows state information to be distributed across multiple nodes within a cluster, reducing the memory burden on each individual node while maintaining local routing efficiency through the cluster's shared data store.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent introduces a shared data store as an intermediary between network nodes and subscriber state information. This shared data store acts as a mediator that all nodes within a cluster can access, eliminating the need for complex routing logic to determine which node holds specific subscriber data, as any node can retrieve it from the shared store.

Inventive Principle:
Principle #24Intermediary (Mediator)

2Adaptability or versatility

If sharding techniques are employed to distribute subscribers across network nodes, then network scalability is improved, but redundancy complexity increases

Engineering Contradiction:
Improvenetwork scalabilityVSAvoidredundancy complexity
Core Design Contradiction:
Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The patent merges the redundancy functionality into the shared data store, which automatically replicates and synchronizes subscriber state information across all nodes within a cluster. This eliminates the need for each node to independently manage redundancy relationships with other nodes, significantly reducing the complexity of providing fault tolerance while maintaining scalability.

Inventive Principle:
Principle #5Merging (Combining)

Solution Approach 2:

The patent uses copying mechanisms where the shared data store maintains copies of subscriber state information across multiple nodes within a cluster. This automatic copying approach simplifies redundancy management compared to sharding, where each node would need to manually establish and maintain redundancy relationships with specific other nodes.

Inventive Principle:
Principle #26Copying

3Ease of operation

If state information is maintained at every network node for every subscriber, then routing simplicity is maintained, but memory requirements become prohibitive

Engineering Contradiction:
Improverouting simplicityVSAvoidmemory requirements
Core Design Contradiction:
Ease of operationVSQuantity of substance

Solution Approach 1:

The patent creates a universal shared data store that serves all network nodes within a cluster, allowing any node to access any subscriber's state information. This multi-functional approach maintains routing simplicity (any node can handle any subscriber's requests) while avoiding the prohibitive memory requirements of duplicating full state information at every node, as the shared store consolidates the data.

Inventive Principle:
Principle #6Universality (Multi-functionality)

4Quantity of substance

If a longstanding responsibility is allocated to each network node for a subset of subscribers, then memory requirements per node are reduced, but the complexity of handling failures and redundancy increases

Engineering Contradiction:
Improvememory requirements per nodeVSAvoidfailure handling complexity
Core Design Contradiction:
Quantity of substanceVSReliability

Solution Approach 1:

The shared data store acts as an intermediary that abstracts away the complexity of failure handling. When a node fails, the shared data store automatically makes the relevant subscriber state information available to other nodes in the cluster, eliminating the need for complex inter-node redundancy protocols and manual failover procedures that would be required in a sharded architecture.

Inventive Principle:
Principle #24Intermediary (Mediator)

Data Source

PatentEP2800320B1Processing data
Publication Date: 2018.12.05 METASWITCH NETWORKS LTD
  • EP2800320B1 patent drawingFigure 1
  • EP2800320B1 patent drawingFigure 2
  • EP2800320B1 patent drawingFigure 3

AI summary

Measures for processing data in a telecommunications network. At a first network node in a cluster of network nodes, a communication session setup request requesting setup of a communication session with a subscriber device in a plurality of subscriber devices is received. In response to receipt of the communication session setup request, registration data relating to the subscriber device is retrieved from a shared registration data store. The communication session setup request is routed according to the retrieved registration data. The retrieved registration data relating to the subscriber device was previously stored in the shared registration data store by a second network node in the cluster of network nodes in relation to a registration procedure conducted for the subscriber device by the second network node, the second network node being a different network node to the first network node in the cluster of network nodes.