Cloud Database Disruption Detection With Proactive Failover
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Multi-tenant database systems in public cloud services face challenges in detecting potential disruptions and failover operations efficiently, leading to non-trivial delays that can result in failure to meet service-level agreement (SLA) metrics, causing revenue loss and client dissatisfaction.
Innovation Solution
A monitoring service in a public cloud computer system continuously tracks metrics from various sources to identify potential disruptions, selects a primary database candidate from standby databases, and triggers a failover operation based on predefined scenarios, reducing the time required to mitigate downtime.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional database disruption detection methods are used, then system stability is maintained, but detection delay increases causing SLA metric failures
Solution Approach 1:
The patent implements preliminary action by continuously monitoring database metrics and pre-identifying potential disruption scenarios before they fully manifest. The system evaluates multiple disruption scenarios in advance and prepares failover candidates proactively, enabling faster response when actual disruptions occur. This pre-positioning of detection and response mechanisms directly reduces detection delay while maintaining reliability.
Solution Approach 2:
The patent employs feedback mechanisms by continuously collecting database metrics, comparing them against disruption scenario thresholds, and automatically triggering failover operations. The monitoring service receives feedback from database performance indicators and adjusts its detection and response actions accordingly, creating a closed-loop system that improves both detection speed and SLA compliance through iterative optimization.
2Measurement precision
If comprehensive monitoring of multiple disruption scenarios is implemented, then detection accuracy improves, but system complexity increases
Solution Approach 1:
The patent applies segmentation by dividing the monitoring system into distinct functional modules: metric collection components, disruption scenario evaluation components, and failover trigger components. Each module handles specific aspects of disruption detection independently, making the overall complex system manageable and maintainable while achieving comprehensive monitoring coverage through coordinated module interactions.
Solution Approach 2:
The patent uses an intermediary approach by introducing a monitoring service that acts as a mediator between the database system and failover execution. This intermediary service consolidates the complexity of monitoring multiple disruption scenarios in a centralized location, simplifying the architecture by avoiding distributed monitoring complexity while maintaining comprehensive detection accuracy through standardized evaluation protocols.
3Productivity
If automated failover operations are implemented, then service continuity is improved, but operational control is reduced
Solution Approach 1:
The patent implements dynamics by designing an adaptive failover system that can adjust its automation level based on disruption severity and operational context. The system dynamically evaluates disruption scenarios and can transition between automated failover and manual approval modes, providing both service continuity through automation and operational control through configurable approval requirements for different disruption types.
Data Source
AI summary
Techniques are disclosed relating to a monitoring service executing in a public cloud computer system. A method may include receiving metrics for a database system implemented on a single instance of a virtual machine in the public cloud computer system. The metrics may include a set of metrics indicative of status of the database system, a set of metrics indicative of status of the virtual machine, and a set of metrics indicative of status of the public cloud computer system. The method may also include continuously determining a primary database candidate from a set of standby databases, and detecting that metrics correspond to one of a plurality of disruption scenarios. The method may further include issuing, based on the detecting, a command to trigger a failover to the primary database candidate.