Tiered Database Storage Segmentation for Replication Cost Reduction

The patent divides the production database into multiple sections (e.g., frequently accessed sections and infrequently accessed sections) and replicates only specific sections to backup databases. This segmentation allows selective replication based on data importance and access patterns, reducing the volume of data that needs to be replicated while maintaining adequate data protection for critical sections.

The patent applies different replication strategies to different sections of the database based on their specific characteristics. Critical sections that require high data protection are replicated to enterprise class storage systems, while less critical sections are replicated to lower quality storage systems or replicated less frequently. This local quality approach optimizes storage costs by matching replication quality to data importance.

The patent segments the database into sections with different retrieval requirements. Frequently accessed sections are replicated to higher quality storage systems that provide better retrieval capabilities, while infrequently accessed sections are replicated to lower quality storage systems. This segmentation ensures that data retrieval capability is optimized for sections where it matters most while reducing costs for sections where retrieval speed is less critical.

The patent applies different storage quality levels to different database sections based on their specific retrieval requirements. Critical sections requiring fast retrieval are stored on enterprise class storage systems, while non-critical sections are stored on lower quality systems. This local quality differentiation resolves the contradiction by ensuring adequate retrieval capability for important data while reducing overall storage costs.

The patent divides the database into sections with different replication frequency requirements. Critical sections are replicated frequently to ensure data protection, while non-critical sections are replicated less frequently. This segmentation reduces the overall replication burden on the system, improving performance while maintaining adequate data protection for sections that require it.

The patent implements different replication frequencies for different database sections based on their characteristics. Frequently changing or critical sections are replicated at higher frequencies, while stable or non-critical sections are replicated at lower frequencies. This periodic action with varying frequencies optimizes the balance between data protection and system performance by replicating only when necessary.