Partial Block Deallocation for SSD Rebuild Optimization

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Conventional storage systems face inefficiencies in data rebuilding due to the need for parity units and increased FRU capacity, leading to longer rebuild times and decreased performance as FRUs degrade.

Innovation Solution

Implementing direct flash memory storage devices to manage storage at a finer granularity, allowing for RAID/parity schemes at a data block level, enabling partial deallocation of uncorrectable data blocks and reducing the amount of data to be rebuilt.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If conventional storage systems use FRU-level RAID stripes, then data redundancy is maintained, but rebuild times increase and performance decreases as FRUs degrade

Engineering Contradiction:
Improvedata redundancyVSAvoidrebuild times
Core Design Contradiction:
ReliabilityVSLoss of time

Solution Approach 1:

The patent segments the RAID stripe from FRU-level to data block-level. Instead of treating entire FRUs as the unit for RAID operations, the system divides data into smaller blocks and distributes them across FRUs. This segmentation allows partial deallocation when blocks become uncorrectable, reducing rebuild scope from entire FRUs to only affected blocks, thereby reducing rebuild time while maintaining data redundancy.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent implements partial deallocation of only the uncorrectable data blocks rather than deallocating entire FRUs. When corruption is detected in specific blocks, the system performs rebuild operations only for those partial blocks, applying partial action to minimize the amount of data that needs to be rewritten, thus reducing rebuild time and wear on storage devices.

Inventive Principle:
Principle #16Partial or excessive action

2Quantity of substance

If FRU capacity increases to compensate for degradation, then storage capacity is maintained, but rebuild times increase and performance decreases

Engineering Contradiction:
Improvestorage capacityVSAvoidperformance
Core Design Contradiction:
Quantity of substanceVSProductivity

Solution Approach 1:

By segmenting storage management to the data block level rather than FRU level, the system can maintain full storage capacity utilization while performing smaller, more efficient rebuild operations. Only the corrupted blocks need to be rebuilt rather than entire FRUs, improving productivity during degradation events while maintaining total storage capacity.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent changes the granularity parameter from FRU-level to data block-level for RAID operations. This parameter change enables the system to work with much smaller units during rebuild operations, reducing the amount of data that must be read, processed, and rewritten, thereby improving performance even as FRUs degrade over time.

Inventive Principle:
Principle #35Parameter changes

3Reliability

If conventional storage systems perform full FRU rebuilds, then data integrity is restored, but storage device longevity decreases due to increased write operations

Engineering Contradiction:
Improvedata integrityVSAvoidstorage device longevity
Core Design Contradiction:
ReliabilityVSDuration of action of stationary object

Solution Approach 1:

The patent applies partial action by rebuilding only the specific corrupted data blocks rather than entire FRUs. This minimizes the number of write operations required to restore data integrity, reducing wear on storage devices and extending their operational lifespan while still achieving complete restoration of data integrity for the affected blocks.

Inventive Principle:
Principle #16Partial or excessive action

Solution Approach 2:

The patent extracts and isolates only the corrupted data blocks from the FRU for rebuild operations. By taking out only the problematic portions rather than treating the entire FRU as a unit, the system minimizes unnecessary write operations to healthy data, thereby reducing overall wear on storage devices while maintaining data integrity.

Inventive Principle:
Principle #2Taking out (Extraction)

Data Source

PatentUS11656961B2Deallocation within a storage system
Publication Date: 2023.05.23 PURE STORAGE INC
  • US11656961B2 patent drawing
  • US11656961B2 patent drawing
  • US11656961B2 patent drawing

AI summary

Failure information associated with a plurality of blocks of a solid-state storage device of a plurality of solid-state storage devices is received. One or more blocks of the plurality of blocks storing uncorrectable data are identified based on the received failure information. A partial deallocation of the one or more blocks of the plurality of blocks is issued, the partial deallocation indicating that the one or more blocks store uncorrectable data. A remedial action associated with the one or more blocks of the plurality of blocks is performed.