NF Instance Re-selection for Data Consistency
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In 5G networks, the current methods for selecting Network Function (NF) instances do not adequately ensure strong data consistency, leading to potential inconsistencies when re-selecting NF instances, especially in geographically distributed scenarios, which can result in different data states being accessed by parallel processes.
Innovation Solution
The solution involves enhancing consumer logic to prioritize NF instances with strong consistency by provisioning additional information, allowing consumers to re-select service instances that support strong consistency, and embedding data consistency information into existing attributes like Locality within the NF profile.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If NF instances are deployed in geographically distributed scenarios with asynchronous replication, then system availability and scalability are improved, but data consistency between parallel processes deteriorates
Solution Approach 1:
The patent changes the parameter of data consistency from a binary state to a spectrum with multiple levels (strong, causal, eventual consistency). By allowing different consistency levels for different data types and operations, the system can achieve both high availability through asynchronous replication and acceptable consistency for critical operations.
Solution Approach 2:
The patent introduces dynamic consistency adjustment where the system can switch between strong consistency mode (synchronous replication) and eventual consistency mode (asynchronous replication) based on operational requirements. This dynamic behavior allows the system to optimize for availability when consistency can be relaxed, and for consistency when required.
2Reliability
If synchronous replication is used to ensure strong data consistency, then data consistency is improved, but system performance and availability deteriorate
Solution Approach 1:
The patent applies different consistency models to different data types and operations based on their specific requirements. Critical data that requires strong consistency uses synchronous replication, while non-critical data uses asynchronous replication. This localized approach ensures data consistency where needed without sacrificing overall system performance.
3Productivity
If multiple NF instances are deployed in the same NF Set with asynchronous replication, then system scalability is improved, but data state divergence between instances increases
Solution Approach 1:
The patent introduces dynamic consistency adjustment where the system can switch between strong consistency mode (synchronous replication) and eventual consistency mode (asynchronous replication) based on operational requirements. This dynamic behavior allows the system to optimize for availability when consistency can be relaxed, and for consistency when required.
Data Source
AI summary
The present invention faces the issues of re-selection of another NF Service instance (103, 201) for a subsequent request (S-520, S-535) with a same data consistency (Consistency 1) as a previously selected NF Service instance (101). To solve this issue, the present invention provides for identifying at configuration, based on the deployment and on the Storage solution for the Session/context data, which instances (101, 103, 201) of the same service type have access with strong consistency to the same data, and provide this information to the consumer (40), so it is allowed to prioritize selection of an alternative NF service instance that supports strong consistency, whenever possible.