Mirror Path Statistics for Storage Recall Optimization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current data backup systems face inefficiencies in migrating and recalling mirrored data between primary and secondary storage systems, particularly when using remote cloud storage, as they often require mirroring data back to the secondary storage server, leading to increased traffic on mirror paths and processing burdens on primary storage servers.
Innovation Solution
The proposed solution involves using mirror path statistics to determine whether mirror paths are degraded and, if so, splitting the recall operation between both primary and secondary storage servers to reduce traffic on degraded paths, thereby avoiding the need to mirror data back to the secondary storage server, and utilizing access statistics to distribute migration tasks between the primary and secondary servers based on access levels.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If data is migrated from primary storage to remote cloud storage and then recalled, then data availability is improved, but traffic on mirror paths increases and processing burden on primary storage server increases
Solution Approach 1:
The patent divides the recall operation into two independent paths: (1) primary storage server recalls data directly from cloud storage, and (2) secondary storage server recalls data directly from cloud storage. This segmentation eliminates the need for data to traverse the mirror path from primary to secondary storage, reducing traffic on degraded mirror paths and distributing processing burden across both servers.
Solution Approach 2:
The patent introduces a new recall architecture that bypasses the traditional mirror path dimension. Instead of recalling data through the primary- secondary storage mirror path, the system enables direct recall from cloud storage to both primary and secondary storage servers simultaneously, creating a parallel dimension for data retrieval that avoids congested paths.
2Stability of the object's composition
If recall operation mirrors data back to secondary storage server, then data consistency is maintained, but traffic on degraded mirror paths increases
Solution Approach 1:
The patent extracts the data consistency assurance function from the mirror path transmission mechanism. Instead of relying on mirror path traffic to maintain consistency between primary and secondary storage, the system independently recalls data to both servers from cloud storage, with each server maintaining its own data integrity, thereby eliminating harmful traffic on degraded paths while preserving consistency.
3Reliability
If primary storage server handles all recall operations, then recall completeness is ensured, but processing burden on primary storage server increases
Solution Approach 1:
The patent segments the recall processing workload by assigning independent recall operations to both primary and secondary storage servers. The secondary storage server handles its own recall from cloud storage, reducing the processing burden on the primary storage server while ensuring complete data restoration across both systems through parallel operations.
Solution Approach 2:
The secondary storage server performs self-service by independently recalling its own data from cloud storage without burdening the primary storage server. Each server manages its own recall operation autonomously, improving overall processing efficiency while ensuring complete data restoration through coordinated parallel operations.
Data Source
AI summary
Provided are a computer program product, system, and method for using mirror path statistics in recalling extents to a primary storage system and a secondary storage system from a third storage system, A recall request is received to recall extents from the third storage that were migrated from the primary storage or the secondary storage to the third storage. A determination is made as to whether mirror paths used to migrate extents from the primary storage to the secondary storage exceed a degradation criteria. Commands are sent to cause the recall extents to be recalled from the third storage to both the primary storage and the secondary storage in response to determining that the mirror paths exceed the degradation criteria.