Geo-Redundant Cloud Server Sync for Selective Disaster Failover
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing cloud orchestrator systems face inefficiencies in managing geo-redundant cloud servers due to tight coupling between operational and disaster recovery sites, leading to resource wastage and cumbersome manual updates, with no efficient options for data synchronization and load-sharing.
Innovation Solution
Implementing geo-redundant operators that synchronize stateful data between operational and disaster recovery sites based on health information and synchronization parameters, enabling efficient data management and load-sharing.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If a load balancer completely shifts operations to the DR site upon identifying a disaster in at least one network component, then disaster recovery capability is improved, but resource utilization deteriorates due to the operational site becoming completely idle
Solution Approach 1:
The patent segments the monolithic failover decision into granular component-level health assessments. Instead of treating the entire operational site as a single unit that must completely fail before switching, the system evaluates individual network components (servers, microservices, databases) independently. This allows selective failover of only the affected components while keeping healthy components operational, thus maintaining resource utilization while ensuring disaster recovery capability.
Solution Approach 2:
The patent applies local quality by enabling different health states and operational modes for different components within the same site. Rather than applying a uniform failover policy to the entire operational site, the system allows some components to remain operational while others are failed over to the DR site. This localized approach optimizes resource utilization by keeping healthy resources active while still providing comprehensive disaster recovery protection.
2Stability of the object's composition
If cloud native micro services are manually updated on both OP and DR sites to eliminate version mismatch, then system consistency is improved, but operational complexity and time consumption worsen
Solution Approach 1:
The patent merges the update management of OP and DR sites into a unified system. The health assessment mechanism and automated configuration synchronization create a single source of truth for microservice versions and configurations. When updates are applied to one site, the system automatically detects the version mismatch and synchronizes the other site, eliminating the need for manual updates at both locations and reducing operational complexity.
Solution Approach 2:
The patent implements feedback mechanisms that continuously monitor microservice versions and health states across OP and DR sites. The load balancer receives health information from both sites and automatically detects version mismatches. This feedback loop triggers automated synchronization processes that update the DR site to match the OP site (or vice versa), eliminating manual intervention and reducing both operational complexity and time consumption while maintaining system consistency.
3Reliability
If the DR site is configured with the same micro services as the OP site, then disaster recovery capability is improved, but resource wastage worsens due to geo-redundancy unawareness
Solution Approach 1:
The patent introduces dynamic configuration management where the DR site's microservice deployment state is no longer static but continuously adapted based on the health and version state of the OP site. The system dynamically synchronizes configurations, enabling the DR site to have a subset of services or different versions based on actual operational needs. This dynamic approach maintains disaster recovery capability while reducing resource wastage by avoiding unnecessary duplication of all services.
Solution Approach 2:
The patent changes the parameter of DR site configuration from fixed (same as OP site) to variable (adaptive based on health information and synchronization parameters). The system allows different deployment states, version configurations, and service subsets at the DR site depending on the operational requirements and health status. This parameter flexibility enables optimized resource allocation while maintaining the ability to recover critical functions during disasters.
4Stability of the object's composition
If unified sync is performed for all stateful data between OP and DR sites, then data consistency is improved, but synchronization complexity and time consumption worsen
Solution Approach 1:
The patent segments the monolithic data synchronization process into granular, component-specific synchronization operations. Instead of synchronizing all stateful data uniformly across the entire system, the health assessment mechanism identifies and synchronizes only the data related to affected components. This segmented approach reduces synchronization scope and time while maintaining consistency for the critical data that actually needs to be synchronized.
Solution Approach 2:
The patent applies partial action by synchronizing only the necessary subset of stateful data rather than performing complete unified synchronization. The health information and synchronization parameters enable the system to identify which specific data elements require synchronization based on component health status and version mismatches. This partial synchronization approach significantly reduces time consumption while maintaining data consistency for the operational requirements.
Data Source
AI summary
Embodiments of the present disclosure disclose a method and a system for managing geo-redundant cloud servers. The method includes: receiving, by a set of first geo-redundant operators associated with a set of first databases hosted on a set of first servers of a first data system, state information of corresponding first server of the set of first servers, health information of a set of second servers of a second data system, from a corresponding first manager of a set of first managers associated with the set of first servers. The method includes synchronizing, by the set of first geo-redundant operators, stateful data from each first database of the set of first databases to a corresponding second database of a set of second databases associated with the set of second servers based on the state information, the health information and one or more synchronization parameters.


