Mirrored Data Consistency Verification via Checksum Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional mirroring and archival backup systems face challenges in ensuring data consistency between source and destination storage systems, leading to potential data loss due to errors, which can exacerbate mirror divergence and require computationally intensive re-establishment of mirroring relationships, often failing to maintain previous point-in-time mirrors.
Innovation Solution
A system and method involving a destination verification module and a source verification module that cooperate to identify the most recent common Persistent Consistency Point Image (PCPI), allowing for verification and correction procedures, including combined verification and repair, and the use of checksums to reduce network bandwidth consumption.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If conventional mirroring systems perform verification and correction procedures, then data consistency between source and destination is improved, but network bandwidth consumption increases due to transmission of checksums and data blocks
Solution Approach 1:
The patent applies partial action by first performing a lightweight checksum verification to identify only the specific data blocks that differ between source and destination. Only these mismatched blocks are then transmitted for correction, rather than transmitting all data blocks. This partial approach significantly reduces network bandwidth consumption while still ensuring complete data consistency verification.
2Reliability
If mirroring systems establish new mirroring relationships to correct errors, then data consistency is restored, but computational overhead and time increase due to re-establishment processes
Solution Approach 1:
The patent implements preliminary action by performing verification of the existing mirroring relationship before errors cause data loss. The system proactively identifies mismatches through checksum verification and corrects them by transmitting only the necessary data blocks. This preliminary verification and correction approach prevents the need for time-consuming complete re-establishment of mirroring relationships.
Solution Approach 2:
The patent applies discarding and recovering by identifying and discarding only the specific corrupted or mismatched data blocks through checksum verification, then recovering them by transmitting only those specific blocks from source to destination. This selective approach is far more efficient than discarding and re-transmitting entire data sets or re-establishing complete mirroring relationships.
3Measurement precision
If verification modules transmit complete data blocks for verification, then verification accuracy is improved, but network bandwidth consumption and transmission time increase
Solution Approach 1:
The patent applies segmentation by dividing the verification process into two distinct stages: first transmitting and verifying only checksums (which are small in size), then transmitting complete data blocks only for those specific blocks that fail the checksum verification. This segmentation allows the system to achieve high verification accuracy through checksums while minimizing network bandwidth consumption by limiting full block transmission to only the necessary cases.
Data Source
AI summary
A system and method verifies and repairs mirrored data sets. Verification modules executing on both the source and destination computers cooperate to identify a common persistent consistency point image (PCPI) of the source and destination data sets. Once the common image has been identified, the source sends a data stream comprising the data from the source data set to the destination to be compared with the destination data set.