Data Destaging via Pairwise Correlation Analysis

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing data destaging methods in information handling systems do not consider data content or disk locations, leading to repeated and unrelated write I/O operations, which increase disk fragmentation and require significant power for future defragmentation.

Innovation Solution

The system determines pairwise correlations between data in non-volatile memory regions and destages data concurrently or sequentially with related data, reducing the need for future defragmentation and power consumption.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If data is destaged without considering data content or disk locations, then the destaging process is simple and fast, but disk fragmentation increases and requires significant power for future defragmentation

Engineering Contradiction:
Improvedestaging speedVSAvoidpower consumption for defragmentation
Core Design Contradiction:
ProductivityVSUse of energy by moving object

Solution Approach 1:

The system performs preliminary analysis of data correlations and disk locations before destaging occurs. By pre-determining which data blocks are correlated and planning destaging operations accordingly, the system avoids future defragmentation needs while maintaining efficient destaging performance

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The system continuously monitors disk fragmentation levels and adjusts destaging operations based on feedback about data correlations and location patterns. This adaptive approach ensures optimal destaging performance while minimizing future defragmentation requirements

Inventive Principle:
Principle #23Feedback

2Device complexity

If data is destaged without considering data content or disk locations, then the destaging process is simple, but disk fragmentation increases requiring future defragmentation work

Engineering Contradiction:
Improvedestaging process complexityVSAvoiddisk fragmentation
Core Design Contradiction:
Device complexityVSObject-generated harmful factors

Solution Approach 1:

The system performs preliminary analysis of data correlations and disk locations before destaging occurs. By pre-determining which data blocks are correlated and planning destaging operations accordingly, the system avoids future defragmentation needs while maintaining efficient destaging performance

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The system merges correlated data blocks into unified destaging operations. By combining multiple related data blocks that should be written to contiguous disk locations, the system reduces the number of separate write operations and minimizes disk fragmentation

Inventive Principle:
Principle #5Merging (Combining)

3Productivity

If repeated and unrelated write I/O operations are performed, then data can be destaged quickly, but disk fragmentation increases leading to more defragmentation work

Engineering Contradiction:
Improvedata destaging throughputVSAvoidtime for defragmentation
Core Design Contradiction:
ProductivityVSLoss of time

Solution Approach 1:

The system performs preliminary analysis of data correlations and disk locations before destaging occurs. By pre-determining which data blocks are correlated and planning destaging operations accordingly, the system avoids future defragmentation needs while maintaining efficient destaging performance

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The system merges correlated data blocks into unified destaging operations. By combining multiple related data blocks that should be written to contiguous disk locations, the system reduces the number of separate write operations and minimizes disk fragmentation

Inventive Principle:
Principle #5Merging (Combining)

Data Source

PatentUS20250200132A1Data defragmentation to reduce power consumption
Publication Date: 2025.06.19 DELL PROD LP
  • US20250200132A1 patent drawing
  • US20250200132A1 patent drawing
  • US20250200132A1 patent drawing

AI summary

An information handling system may include at least one processor; a volatile memory; and a non-volatile memory. The information handling system may be configured to: determine pairwise correlations between data in a plurality of regions of the non-volatile memory; and destage data from the volatile memory to the non-volatile memory in accordance with the pairwise correlations, such that first data is destaged with second data having a high degree of correlation to the first data.