Transaction Processing System with Backup Server Failover
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Modern business application systems face challenges in maintaining transaction processing continuity due to system failures, as existing failure-detection and take-over mechanisms require significant time to detect and respond to server failures, leading to prolonged downtime and inability to hide failures from users.
Innovation Solution
A transaction processing method that utilizes a distributed agreement protocol and exclusive control techniques to allow transactions to be resent to a backup server without waiting for failure detection, ensuring that only one server farm has the processing authority token at a time, and uses data replication and consistency techniques to ensure transactions commit on the latest data.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If failure-detection and take-over-processing mechanisms are implemented, then system reliability is improved, but the response time increases significantly (10 seconds to several minutes)
Solution Approach 1:
The patent applies preliminary action by pre-configuring backup servers and pre-establishing failover capabilities before failures occur. The system maintains standby servers with pre-loaded data and configurations, enabling immediate takeover when a failure is detected, thus reducing the 10-second to several-minute delay inherent in traditional failure-detection mechanisms.
Solution Approach 2:
The patent introduces an intermediary load balancer or traffic management component that can redirect traffic to backup servers without requiring direct failure detection between client and server. This intermediary layer absorbs the failure detection and routing decisions, allowing the actual server failure to be handled silently in the background while maintaining service continuity.
2Measurement precision
If message exchange attempts are increased to improve failure detection accuracy, then detection precision is improved, but the time required for failure detection increases
Solution Approach 1:
The patent uses copying by maintaining duplicate server instances (primary and backup servers) with identical data and configuration. Instead of repeatedly attempting message exchange with the same server to detect failure, the system can immediately switch to a copied backup instance, achieving both high detection accuracy and rapid response by validating failure through the copy's ability to assume the primary role.
3Speed
If the reply waiting time is set too small, then the response speed is improved, but false failure detection occurs
Solution Approach 1:
The patent applies local quality by implementing different waiting time thresholds for different types of operations or different server pairs. Critical operations may use shorter waiting times with aggressive failover, while non-critical operations use longer waiting times to avoid false positives. This localized differentiation allows the system to optimize for speed where necessary while maintaining reliability where appropriate.
Data Source
AI summary
The present invention proposes a system for resending a process to a backup server farm from a client without waiting for the failure-detection, if no reply is received for a certain time. The transaction processing mechanism of the present invention has a transaction start processing mechanism in which an exclusive control using a valid processing authority token and data consistency are combined, and a commit processing mechanism in which determination on whether a commit is available or not based on a distributed agreement and replication of updated data. With the mechanisms, a system for shortening a service halt time when a failure occurs to a time as such it appears to a client that the service does not stop is provided.