Tiered Data Storage Deduplication via Content-Addressable Hashing

The patent merges multiple versions of the same data into a single stored copy by implementing a content-addressable storage system. When data is written, the system checks if identical data already exists using hashing; if found, it stores only a reference to the existing copy rather than duplicating the data. This combining approach maintains data resiliency through versioning metadata while eliminating redundant storage of identical data blocks.

The patent uses content-addressable copying where data is copied only when actually modified. The system creates snapshots and versions by copying only the changed data blocks rather than entire datasets. This selective copying mechanism, combined with deduplication, reduces storage requirements while maintaining the ability to restore any previous version of data.

The patent implements self-service mechanisms where the storage system automatically manages versioning and deduplication without requiring complex external management. The content-addressable storage automatically hashes incoming data, checks for existing copies, and manages version metadata. This automation reduces computational overhead by eliminating manual version tracking and location management while maintaining data accessibility.

The patent changes the fundamental parameter of data identification from location-based to content-based addressing. By using hashing of data content as the address key, the system transforms how data is located and versioned. This parameter change simplifies the management of multiple versions since each unique content automatically gets a unique identifier, eliminating the need for complex version tracking infrastructure.

The patent segments data into discrete blocks that can be independently addressed, stored, and managed across different locations (local and cloud). Each data block is hashed and stored as a separate entity, allowing the system to distribute segments across tiered storage locations while eliminating redundancy. This segmentation enables efficient deduplication where identical blocks anywhere in the system are recognized and deduplicated.

The patent creates a universal content-addressable storage layer that operates consistently across local and cloud-based systems. The same deduplication and versioning mechanisms work uniformly regardless of whether data resides locally or in the cloud. This universal approach eliminates redundant data across the entire distributed system while maintaining resiliency through consistent data protection policies across all storage locations.