Standby Cache Recovery Using Failover Log Access Probability
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing disaster recovery systems are inefficient in minimizing data loss and business disruption during catastrophic events, leading to significant financial losses due to infrastructure and application failures.
Innovation Solution
An Application Recovery Accelerator system that detects cache metrics based on the probability of data object access in failover logs, caching data objects in a standby cache using a cache metric to optimize recovery processes.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Speed
If data objects are cached in standby cache during disaster recovery, then recovery speed is improved, but system complexity and resource consumption increase
Solution Approach 1:
The system performs preliminary actions by analyzing failover logs before disasters occur, identifying frequently accessed data objects and pre-caching them in standby cache. This proactive approach ensures that when a disaster strikes, the cached data is immediately available, significantly reducing recovery time without requiring complex real-time decision-making during the disaster event.
Solution Approach 2:
The system segments the failover log into time segments and analyzes access patterns within each segment. By dividing the log analysis into manageable time-based portions, the system can efficiently identify hot data objects without overwhelming computational resources, balancing recovery speed improvement with acceptable system complexity.
2Reliability
If all data objects are cached in standby cache, then data loss is minimized, but memory consumption increases
Solution Approach 1:
Instead of uniformly caching all data objects, the system applies local quality by differentiating between frequently accessed (hot) data objects and less frequently accessed data. By analyzing access patterns in failover logs and caching only the hot data objects in standby cache, the system minimizes memory consumption while still providing robust protection against data loss for the most critical data.
Solution Approach 2:
The system implements partial action by caching only a subset of data objects—specifically those identified as hot data through log analysis—rather than caching all data objects. This selective approach achieves sufficient data loss prevention for critical operations while maintaining acceptable memory consumption levels.
3Measurement precision
If failover logs are analyzed in detail to determine cache metrics, then caching accuracy is improved, but processing time increases
Solution Approach 1:
The system segments the failover log into multiple time segments and analyzes access patterns within each segment separately. This segmentation allows for detailed and accurate analysis of caching metrics without requiring the system to process the entire log as a single monolithic structure, thereby maintaining high caching accuracy while reducing overall processing time through divide-and-conquer methodology.
Data Source
AI summary
An embodiment includes detecting, by a Recovery System, a cache metric where the cache metric comprises a probability of access to a data object for each of a plurality of time segments of a failover log, where the probability of access to the data object for each of the plurality of time segments is based on an access time of the data object contained in the failover log. The embodiment also includes responsive to receiving the cache metric, caching by the Recovery System of the data object based on the cache metric where the data object is cached in a standby cache of the Recovery System.