Autonomous Data Sharding for Disaster Recovery RTO Compliance
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current disaster recovery solutions lack an autonomous mechanism to optimize both recovery point objective (RPO) and recovery time objective (RTO) in multi-site configurations, especially for large data sets with non-uniform storage capacity across data centers.
Innovation Solution
A computer-implemented method and system for data sharding and topology alterations that estimates the time needed to copy a data store, determines if it exceeds the RTO, creates data shards if necessary, and alters the data store topology by copying these shards to other data centers, ensuring that the RTO is met in the event of a data center going offline.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If asynchronous replication strategies are used to replicate data across multiple data centers, then data availability and redundancy are improved, but the RPO is limited by network speed and the amount of data that needs to be migrated
Solution Approach 1:
The patent segments the data store into multiple data shards that can be independently replicated. By dividing the large data set into smaller manageable pieces, the system can parallelize the replication process across multiple data centers, significantly reducing the time required to achieve the RPO while maintaining data availability and redundancy.
2Reliability
If a second identical live data center is deployed for active-active configuration, then recovery capability is improved, but the complexity of managing multiple data centers with non-uniform storage capacity increases
Solution Approach 1:
The patent implements a dynamic topology management system that automatically adjusts data shard distribution across data centers based on real-time factors including network conditions, storage capacity availability, and failure scenarios. This dynamic approach allows the system to adapt to non-uniform storage capacities and optimize replication paths without manual intervention, reducing operational complexity while maintaining recovery capability.
Solution Approach 2:
The system performs autonomous data sharding and topology optimization without requiring manual administration. The autonomous mechanism evaluates RTO requirements, determines optimal shard distributions, and executes replication strategies automatically, eliminating the need for complex manual topology management while ensuring recovery capabilities meet SLA requirements.
3Reliability
If autonomous data sharding is implemented to meet strict RTO and RPO requirements, then disaster recovery performance is improved, but the automation mechanism complexity increases
Solution Approach 1:
The autonomous data sharding mechanism incorporates feedback loops that continuously monitor network conditions, storage capacity, and replication progress. Based on this feedback, the system dynamically adjusts shard sizes, replication priorities, and data distribution strategies to meet RTO and RPO requirements. This feedback-driven approach enables complex automated decision-making that adapts to changing conditions while maintaining recoverability.
Data Source
AI summary
A computer-implemented method, a computer program product, and a computer system for data sharding and topology alterations for disaster recovery preparation. A computer system estimates time needed to copy a data store of a data center that is assumed offline. A computer system determines whether the time needed is greater than a recovery time objective (RTO). A computer system creates a data shard for the data store, in response to determining that the time needed is greater than the RTO. Prior to the data center being offline, a computer system alters a current topology of the data store in a network connecting multiple data centers, by copying the data shard to a destination data center that does not contain the data store in the current topology, where, with an altered topology, the RTO is met in an event of the data center being offline.