Tiered Storage Defragmentation via Weighted Flash Factor

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Tiered data storage systems face inefficiencies due to indiscriminate data placement, leading to increased fragmentation and reduced lifespan of storage media, particularly in solid-state drives (SSDs) where data is dispersed and requires unnecessary erase cycles, resulting in higher access times and reduced device longevity.

Innovation Solution

A computer-implemented method and system that calculates a weighted flash fragmentation factor (WFFF) to assess the severity of fragmentation, which is shared with the host device for informed tier selection, allowing data to be stored in optimal tiers based on access frequency and fragmentation impact, thereby reducing fragmentation and extending SSD lifespan.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If data is written indiscriminately to storage tiers without considering fragmentation impact, then storage capacity is utilized, but data fragmentation increases and access times worsen

Engineering Contradiction:
Improvestorage capacity utilizationVSAvoiddata access time
Core Design Contradiction:
ProductivityVSLoss of time

Solution Approach 1:

The system performs preliminary calculation of the weighted flash fragmentation factor before data is written to storage tiers. By assessing the potential fragmentation impact in advance and sharing this information with the host device, the system enables informed tier selection that prevents future fragmentation issues, thereby maintaining both storage utilization and access performance.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The system implements a feedback mechanism where the calculated weighted flash fragmentation factor is shared with the host device. This feedback loop allows the host to adjust its tier selection strategy based on real-time fragmentation assessments, creating a closed-loop system that continuously optimizes data placement to balance capacity utilization with access time performance.

Inventive Principle:
Principle #23Feedback

2Productivity

If data is written to SSD tiers, then storage capacity is utilized, but device lifespan is reduced due to unnecessary erase cycles

Engineering Contradiction:
Improvestorage capacity utilizationVSAvoidSSD lifespan
Core Design Contradiction:
ProductivityVSDuration of action of stationary object

Solution Approach 1:

The system calculates the weighted flash fragmentation factor before data writing operations to predict potential fragmentation and its impact on SSD lifespan. By performing this assessment in advance and communicating it to the host device, the system enables proactive tier selection that avoids placements likely to cause excessive fragmentation and unnecessary erase cycles, thereby extending SSD lifespan while maintaining storage utilization.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The feedback mechanism shares fragmentation risk information with the host device, enabling the host to make informed decisions about tier selection. This feedback loop allows the system to balance storage capacity utilization with SSD lifespan by directing data to tiers that minimize fragmentation-induced erase cycles, thus preserving SSD longevity.

Inventive Principle:
Principle #23Feedback

3Productivity

If data is fragmented across storage media, then storage capacity is utilized, but access times increase due to non-sequential access requirements

Engineering Contradiction:
Improvestorage capacity utilizationVSAvoiddata access speed
Core Design Contradiction:
ProductivityVSSpeed

Solution Approach 1:

The system performs preliminary assessment of the weighted flash fragmentation factor to predict how data placement will affect fragmentation levels. By calculating this factor before writing operations and sharing it with the host device, the system enables tier selection that anticipates and prevents fragmentation, ensuring data remains accessible in sequential blocks and maintaining high access speeds while utilizing storage capacity effectively.

Inventive Principle:
Principle #10Preliminary action

4Loss of time

If defragmentation operations are performed on SSDs, then data access times are improved, but device lifespan is reduced due to additional erase cycles

Engineering Contradiction:
Improvedata access timeVSAvoidSSD lifespan
Core Design Contradiction:
Loss of timeVSDuration of action of stationary object

Solution Approach 1:

The system applies preliminary anti-action by preventing fragmentation in the first place through intelligent tier selection based on the weighted flash fragmentation factor. By assessing fragmentation risk before data writing and guiding host device decisions, the system avoids creating fragmented data that would later require defragmentation operations, thereby eliminating the trade-off between access time improvement and lifespan reduction.

Inventive Principle:
Principle #9Preliminary anti-action

Solution Approach 2:

The system performs preliminary fragmentation assessment and takes preventive action by directing data to appropriate tiers before fragmentation occurs. This proactive approach eliminates the need for subsequent defragmentation operations, preserving SSD lifespan while maintaining optimal access times through proper initial data placement.

Inventive Principle:
Principle #10Preliminary action

Data Source

PatentUS11662932B2Tiered storage system with defragmentation based on weighted flash fragmentation factor
Publication Date: 2023.05.30 SANDISK TECHNOLOGIES LLC
  • US11662932B2 patent drawing
  • US11662932B2 patent drawing
  • US11662932B2 patent drawing

AI summary

Systems and method for providing tier selection for data based on a weighted flash fragmentation factor. A weighted flash fragmentation factor is determined indicating a severity of fragmentation in a non-volatile storage based on a logical block address range in a logical-to-physical mapping table for data from a host device to be stored in the tiered data storage system. The factor is shared with the host device to determine a tier selection. The data is stored according to the tier selection based on the factor.