Snapshot Cross-Validation for Distributed Data Corruption Detection

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Distributed databases face challenges in timely detection, root cause analysis, and mitigation of data corruption, which can lead to invalid query results and database unavailability.

Innovation Solution

The system employs a method to detect data corruption by cross-validating snapshots between replicas in a distributed data set system, using log sequence numbers to identify inconsistencies and generating alerts for administrators.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If distributed database systems use multiple nodes for scaling and high concurrency, then productivity and scalability are improved, but the probability of data corruption increases due to increased coordination requirements among components

Engineering Contradiction:
Improvethroughput and concurrencyVSAvoiddata corruption risk
Core Design Contradiction:
ProductivityVSReliability

Solution Approach 1:

The system performs preliminary actions by generating snapshots at predetermined intervals and cross-validating them before data corruption can propagate. The validation process proactively compares log sequence numbers between primary and replica nodes to detect inconsistencies early, preventing corrupted data from affecting query results and maintaining system reliability despite high concurrency operations

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The system implements feedback mechanisms by continuously monitoring log sequence numbers between primary and replica nodes through snapshot cross-validation. When inconsistencies are detected, the system generates alerts and can initiate recovery procedures, creating a closed-loop control system that maintains data integrity across distributed nodes even during high-volume transactions

Inventive Principle:
Principle #23Feedback

2Measurement precision

If the system implements snapshot cross-validation between replicas, then data corruption detection capability is improved, but device complexity increases due to additional coordination and validation operations

Engineering Contradiction:
Improvedata corruption detection accuracyVSAvoidvalidation system complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The system extracts the validation function into a separate, dedicated process that independently compares snapshots without interfering with primary database operations. By separating the validation logic from the main transaction processing path, the system achieves accurate corruption detection while minimizing the complexity impact on core database functions

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

The system changes parameters by using lightweight metadata comparisons (log sequence numbers) rather than full data content validation. This parameter-based approach maintains high detection accuracy by identifying inconsistencies through sequence number mismatches, while significantly reducing the computational complexity and resource requirements compared to complete data comparison

Inventive Principle:
Principle #35Parameter changes

3Reliability

If snapshots are generated at predetermined intervals, then the time window for operating on incorrect data is reduced, but loss of time occurs due to the validation process itself

Engineering Contradiction:
Improveintegrity of query resultsVSAvoidvalidation processing time
Core Design Contradiction:
ReliabilityVSLoss of time

Solution Approach 1:

The system performs partial validation by comparing only critical metadata elements (log sequence numbers) rather than validating entire data sets. This partial action approach sufficiently detects data corruption to maintain query result integrity while minimizing the time consumed by validation operations, as sequence number comparison is computationally inexpensive compared to full data verification

Inventive Principle:
Principle #16Partial or excessive action

Data Source

PatentEP4327208B1Snapshot-based data corruption detection
Publication Date: 2025.05.21 MICROSOFT TECHNOLOGY LICENSING LLC
  • EP4327208B1 patent drawingFigure 1
  • EP4327208B1 patent drawingFigure 2~3
  • EP4327208B1 patent drawingFigure 4

AI summary

Embodiments described herein detect data corruption in a distributed data set system. For example, a system comprises node(s) for processing queries with respect to a distributed data set comprising a plurality of storage segments. A write transaction resulting from a query with respect to a particular storage segment is logged in a log record that describes a modification to the storage segment. A log service provides the log record to a data server managing a portion of the distributed data set in which the storage segment is included, which performs the write transaction with respect to the storage segment. For redundancy purposes, the data server has replica(s) that manage respective replicas of the portion of the distributed data set managed thereby. For backup purposes, snapshots of the replica(s) are periodically generated. To determine a data corruption, a snapshot of one replica is cross-validated with a snapshot of another replica.