Integration Server Failover Using Queued Flow State Recovery
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing integration server systems fail to maintain uninterrupted data integration flows due to crashes, requiring restarts from the beginning, leading to inefficiencies and wasted processing.
Innovation Solution
Implementing a backup integration server that can take over as a primary server upon failure, utilizing a queuing system to manage state messages and non-deterministic data, allowing seamless continuation of integration flows without restarts.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If a single primary integration server processes integration flows, then the system structure is simple, but the system reliability deteriorates when the server fails
Solution Approach 1:
The patent creates a backup integration server that copies the state and configuration of the primary server. This backup server remains idle until needed, providing fault tolerance without permanently doubling the active processing capacity. The copying mechanism ensures that when failure occurs, the backup can immediately take over with minimal disruption.
Solution Approach 2:
The system performs preliminary actions by pre-configuring a backup server with the necessary state information before failure occurs. The backup server maintains a copy of the integration flow state in advance, so when the primary server fails, the transition is seamless and no work needs to be restarted from scratch.
2Productivity
If integration flows are restarted from the beginning after server failure, then system simplicity is maintained, but productivity deteriorates due to wasted processing
Solution Approach 1:
The backup server performs preliminary state capture by maintaining copies of integration flow state information while the primary server is operational. This preliminary action ensures that when failure occurs, the backup can resume processing immediately without restarting from the beginning, thereby preventing both time loss and productivity deterioration.
Solution Approach 2:
The system ensures continuity of useful action by having the backup server take over seamlessly when the primary fails. The integration flows continue processing without interruption or restart, maintaining continuous productive action. The state transfer mechanism ensures that the backup server picks up exactly where the primary left off, preserving the continuity of the integration process.
3Reliability
If a backup integration server is implemented, then system reliability improves, but device complexity increases
Solution Approach 1:
Rather than implementing complex redundant active systems, the patent uses a copying approach where the backup server maintains state information as a copy of the primary. This reduces complexity compared to active-active configurations while still providing fault tolerance. The copy mechanism is simpler than full replication because the backup is passive and only needs to maintain state, not actively process flows.
Solution Approach 2:
The patent introduces a state transfer mechanism as an intermediary between the primary and backup servers. This intermediary handles the complexity of state synchronization and server transition, isolating the complexity from the core integration logic. The mediator pattern allows the backup server to receive state information without requiring complex direct communication protocols with the primary server.
Data Source
AI summary
A computer-implemented method for retrieving, by a second integration server from a queuing system, a first state message comprising a flow correlation identification (ID). The method further includes initializing, by the second integration server, an integration flow based on the flow correlation ID and processing, by the second integration server, the integration flow as a backup integration server to a first integration server. In response to determining that the first integration server has failed, the method further includes processing, by the second integration server, the integration flow as a primary integration server.