Database Replica Recovery with Delta Storage Computers

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

State-of-the-art database replica recovery is inefficient due to sequential data synchronization, leading to increased network and processing load, and unbounded waste of resources during recovery, which affects OLTP latency and disk lifespan.

Innovation Solution

Implementing a delta storage computer that persistently stores only modified database blocks, acting as a proxy for the failed storage computer, to accelerate recovery by avoiding content interpretation and optimizing network transmission.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If sequential data synchronization is used for database replica recovery, then data consistency is maintained, but recovery time increases and network load increases

Engineering Contradiction:

Improvedata consistencyVSAvoidrecovery time

Core Design Contradiction:

ReliabilityVSLoss of time

Solution Approach 1:

The patent segments the synchronization data into change vectors that track only modified database blocks. Instead of transmitting and replaying complete sequential data streams, the system divides the recovery process into discrete block-level changes, allowing parallel processing and reducing overall recovery time while maintaining data consistency through version vectors.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The system performs preliminary actions by maintaining change vectors and version information during normal operation. When a replica fails, the recovery process can immediately apply pre-computed change vectors without needing to reprocess entire data streams, significantly reducing recovery time while ensuring consistency through the pre-established version ordering.

Inventive Principle:

Principle #10Preliminary action

2Reliability

If sequential replay of change vectors is performed, then data consistency is ensured, but processing load and network transmission time increase

Engineering Contradiction:

Improvedata consistencyVSAvoidprocessing throughput

Core Design Contradiction:

ReliabilityVSProductivity

Solution Approach 1:

The patent segments the recovery workload into independent block-level operations using change vectors. Each change vector represents a discrete modification that can be applied independently, enabling parallel processing across multiple cores or nodes while maintaining consistency through version vector ordering, thus increasing processing throughput without sacrificing reliability.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The system adds a version vector dimension to the traditional sequential replay approach. By incorporating version vectors that encode causal relationships and timestamps, the system enables parallel execution in the time dimension while maintaining consistency through the version ordering dimension, effectively transforming a one-dimensional sequential process into a multi-dimensional parallelizable process.

Inventive Principle:

Principle #17Another dimension (Dimensionality change)

3Reliability

If complete data transmission is used for replica recovery, then recovery accuracy is maximized, but network bandwidth consumption increases

Engineering Contradiction:

Improverecovery accuracyVSAvoidnetwork bandwidth

Core Design Contradiction:

ReliabilityVSLoss of energy

Solution Approach 1:

The patent extracts only the essential change information needed for recovery by maintaining change vectors that record modifications to database blocks. Instead of transmitting complete data sets or extensive metadata, the system extracts and transmits only the minimal necessary change information, reducing network bandwidth consumption while preserving recovery accuracy through the structured change vector format.

Inventive Principle:

Principle #2Taking out (Extraction)

4Reliability

If interpreted replay of redo entries is performed, then data consistency is maintained, but disk write operations increase reducing disk lifespan

Engineering Contradiction:

Improvedata consistencyVSAvoiddisk lifespan

Core Design Contradiction:

ReliabilityVSDuration of action of stationary object

Solution Approach 1:

The system uses change vectors as lightweight copies of modification information rather than replaying complete redo entries. The change vectors contain essential modification data that can be applied directly to database blocks without requiring extensive disk write operations, reducing wear on persistence media while maintaining consistency through the version vector ordering mechanism.

Inventive Principle:

Principle #26Copying

Data Source

PatentUS20250335462A1Storage failure handling and rebalance in database aware distributed data store

Publication Date: 2025.10.30 ORACLE INT CORP

AI summary

For database high availability and for accelerated recovery of a failed replica of a database, a storage computer is dynamically allocated and temporarily persists database content modifications until the database replica is ready to receive the modifications. The storage computer is not allocated storage that stores the database. The storage computer persists a recent portion of the database and later receives a request to synchronize the recovering replica. During recovery, the storage computer responsively sends the portion of the database to the recovering replica. For acceleration, recovery herein does not entail content interpretation such as replay of a redo log. For horizontally scaled acceleration involving two distinct storage computers per recovering replica, multiple replicas are concurrently recovered by respective storage computers that each receives recovered database content only from a respective distinct other storage computer.