Near-Synchronous Object Store Replication via Segmented Metadata Maps
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional data replication methods for object stores lack cross-region protection, leading to data loss in case of datacenter failures, and existing near-synchronous models are inefficient due to high storage and latency overheads.
Innovation Solution
A near-synchronous replication system using separate database maps for object information and replication information, with atomic writes to reduce latency and storage overhead, ensuring data protection across geographic regions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If conventional data replication methods are used for object stores, then data protection is provided, but storage overhead and latency are high
Solution Approach 1:
The patent segments the metadata storage into separate database maps: one map stores object information metadata entries, while another map stores replication information metadata entries. This segmentation allows independent optimization of each data structure, enabling efficient retrieval of object information without being burdened by replication state data, thereby reducing latency while maintaining data protection across clusters.
Solution Approach 2:
The patent extracts replication state information from the main object metadata storage and places it in a separate replication information metadata entry. This extraction removes the overhead of managing replication state within the primary object data structure, reducing storage overhead and improving access performance for both object operations and replication operations.
2Reliability
If conventional data replication methods are used for object stores, then data protection is provided, but storage overhead is high
Solution Approach 1:
The patent segments metadata into separate database maps for object information and replication information. This segmentation eliminates redundant storage of replication state within object metadata, reducing overall storage overhead while maintaining complete data protection through the separate tracking of replication states across clusters.
Solution Approach 2:
The patent extracts replication state information into separate replication information metadata entries, removing unnecessary duplication of data in the primary object storage structure. This extraction reduces the quantity of data stored while preserving data protection capabilities through dedicated replication tracking.
3Loss of time
If separate database maps are used for object information and replication information, then latency is reduced, but device complexity increases
Solution Approach 1:
The patent segments metadata storage into separate database maps, which reduces latency by enabling independent access to object information without processing replication state data. The increased complexity is managed through a systematic approach where each map has a dedicated purpose, simplifying the access patterns and making the system easier to manage despite the additional structural elements.
Data Source
AI summary
An illustrative embodiment disclosed herein is an apparatus including a processor having programmed instructions to detect an operation associated with a first object instance in a first cluster, determine that the first object instance is configured for replication to a second cluster, create an object information (info) metadata entry for the first object instance, create an object replication metadata entry for the first object instance, and simultaneously write the object info metadata entry and the object replication metadata entry to a first data structure and a second data structure, respectively.