Intelligent Demand-Driven Data Replication for Cloud ERP Integration
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In modern microservices architectures, existing data replication methods are inefficient due to the need for full data replication, which is impractical for large datasets and can cause latency and resilience issues, especially when integrating with large Enterprise Resource Planning (ERP) systems and cloud-based Software-as-a-Service (SaaS) solutions.
Innovation Solution
Implementing intelligent demand-driven data replication, where only required data is replicated and stored for as long as needed, allowing for asynchronous data retrieval and periodic maintenance to minimize storage and network load, while maintaining resilience and response time.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If full data replication is performed from ERP system to cloud-based application, then data availability is improved, but database storage costs and network traffic increase significantly
Solution Approach 1:
The patent extracts and replicates only the specific master data subsets that are actually needed by the cloud-based application, rather than performing full data replication. This selective extraction approach maintains data availability for required data while significantly reducing the quantity of data stored and transmitted, thereby lowering storage costs and network traffic.
Solution Approach 2:
The patent implements partial replication by replicating only a portion of the total data (specifically, the master data referenced in event data) rather than all data. This partial action approach achieves sufficient data availability for application needs without the excessive storage and network resources required for complete data replication.
2Reliability
If full data replication is performed, then data completeness is improved, but network traffic and processing time increase
Solution Approach 1:
The system extracts only the necessary master data identifiers from event data and replicates only those specific data subsets from the ERP system. This selective extraction eliminates the need to process and transmit unnecessary data, significantly reducing network traffic and processing time while maintaining data completeness for the required data elements.
Solution Approach 2:
The patent applies partial replication by transferring only the subset of master data that is actually referenced and needed by the application, rather than performing excessive full data replication. This approach achieves adequate data completeness for operational needs while minimizing processing time and network resource consumption.
3Stability of the object's composition
If synchronized replication with tight coherence is used, then data consistency is improved, but latency increases
Solution Approach 1:
The patent extracts and replicates only specific master data subsets rather than all data, which reduces the volume of data that needs to be synchronized. This selective approach maintains data consistency for the replicated data while reducing the latency associated with transmitting and processing larger data sets during synchronization events.
4Loss of energy
If demand-driven replication is used, then network traffic is reduced, but data may not be available when needed
Solution Approach 1:
The patent performs preliminary action by proactively replicating required master data from the ERP system to the cloud-based application before it is actually needed. The system identifies which master data is referenced in event data and replicates it in advance, ensuring data availability when needed while avoiding the network traffic of on-demand retrieval. This anticipatory replication strategy eliminates the trade-off between network traffic reduction and data availability.
Data Source
AI summary
In an example embodiment, a solution is provided that allows for intelligent demand-driven data replication. Only the data that is needed has to be replicated, and this data can be kept only as long as needed. This allows cloud-based applications to be built that integrate deeper with the existing customer landscape with the most efficient data footprint. It allows data types that could not be used before to be integrated into an application. There are additional savings due to the reduced data footprint in the database, compared to a full data replication, without giving up resilience or response time.