Algebraic Signatures for Mass Storage Parity Verification

Solution Overview

Problem

Current network storage systems face inefficiencies in verifying data integrity and parity consistency due to the time-consuming nature of parity scrubs, which can take months to complete and significantly impact user I/O and interconnect bandwidth, especially as drive capacities increase.

Innovation Solution

The use of algebraic signatures for data blocks, specifically retrieving and combining row and diagonal parity algebraic signatures to verify data integrity and parity consistency, reducing the data transfer required and allowing for continuous background scrubbing without excessive bandwidth consumption.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If traditional parity scrub methods are used to verify data integrity, then verification completeness is improved, but bandwidth consumption increases and verification time becomes excessively long

Engineering Contradiction:

Improvedata integrity verificationVSAvoidbandwidth consumption

Core Design Contradiction:

ReliabilityVSLoss of energy

Solution Approach 1:

The patent extracts only the essential verification information (algebraic signatures) from the complete data blocks. Instead of reading and verifying entire data blocks during parity scrubs, the system extracts and stores compact algebraic signatures that represent the essential integrity information, allowing verification without transferring full block data across the interconnect.

Inventive Principle:

Principle #2Taking out (Extraction)

Solution Approach 2:

The patent changes the verification parameter from complete data blocks to algebraic signatures. By transforming the verification task from checking entire data blocks to checking compact algebraic signature values, the system maintains verification reliability while dramatically reducing the amount of data that must be transferred during scrub operations.

Inventive Principle:

Principle #35Parameter changes

2Reliability

If traditional parity scrub methods are used to verify data integrity, then verification completeness is improved, but verification time becomes excessively long

Engineering Contradiction:

Improvedata integrity verificationVSAvoidverification time

Core Design Contradiction:

ReliabilityVSLoss of time

Solution Approach 1:

The patent extracts only the essential verification information (algebraic signatures) from the complete data blocks. Instead of reading and verifying entire data blocks during parity scrubs, the system extracts and stores compact algebraic signatures that represent the essential integrity information, allowing verification without transferring full block data across the interconnect.

Inventive Principle:

Principle #2Taking out (Extraction)

Solution Approach 2:

The patent changes the verification parameter from complete data blocks to algebraic signatures. By transforming the verification task from checking entire data blocks to checking compact algebraic signature values, the system maintains verification reliability while dramatically reducing the amount of data that must be transferred during scrub operations.

Inventive Principle:

Principle #35Parameter changes

3Ease of operation

If parity scrubbing is throttled to reduce impact on user I/O, then user I/O performance is improved, but verification throughput decreases

Engineering Contradiction:

Improveuser I/O performanceVSAvoidverification throughput

Core Design Contradiction:

Ease of operationVSProductivity

Solution Approach 1:

The patent extracts only the essential verification information (algebraic signatures) from the complete data blocks. Instead of reading and verifying entire data blocks during parity scrubs, the system extracts and stores compact algebraic signatures that represent the essential integrity information, allowing verification without transferring full block data across the interconnect.

Inventive Principle:

Principle #2Taking out (Extraction)

Data Source

PatentUS8327250B1Data integrity and parity consistency verification

Publication Date: 2012.12.04 NETAPP INC

AI summary

Verifying data integrity and parity consistency of data blocks in an array of mass storage devices includes retrieving a row parity algebraic signature and a diagonal parity algebraic signature for one or more data blocks, a row parity block and a diagonal parity block. The row parity algebraic signatures of the one or more data blocks are logically combined to generate a first result and the first result is compared to the retrieved row parity algebraic signature for the row parity block. The diagonal parity algebraic signatures of the one or more data blocks and the row parity block are logically combined to generate a second result and the second result is compared to the retrieved diagonal parity algebraic signature for the diagonal parity block.