Filesystem Block Grouping for Small-File Deduplication

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Conventional fixed-block-size filesystems waste storage space when storing files or metadata smaller than the fixed block size, leading to inefficiencies in data storage capacity utilization.

Innovation Solution

Implementing a reference count system in a filesystem free space bitmap to group multiple small data files and metadata structures into a single filesystem block, enabling deduplication and reducing the number of used blocks, thereby improving storage density and efficiency.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Ease of manufacture

If a fixed-block-size filesystem is used to store files and metadata, then storage allocation is simplified and management is easier, but storage space is wasted when files or metadata are smaller than the fixed block size

Engineering Contradiction:
Improveease of storage allocationVSAvoidwasted storage space
Core Design Contradiction:
Ease of manufactureVSLoss of substance

Solution Approach 1:

The patent merges multiple small files and metadata structures into a single filesystem block by grouping them together. The filesystem allocator groups multiple small objects (files or metadata) that would individually not fill a block, and stores them together in one block, thereby eliminating the wasted space that would occur if each object were stored in its own separate block.

Inventive Principle:
Principle #5Merging (Combining)

2Ease of operation

If multiple small files are stored separately in individual filesystem blocks, then file access is simpler and more direct, but the number of used blocks increases and storage density decreases

Engineering Contradiction:
Improvefile access simplicityVSAvoidnumber of used blocks
Core Design Contradiction:
Ease of operationVSQuantity of substance

Solution Approach 1:

The patent combines multiple small files into a single filesystem block, reducing the total number of blocks required. The filesystem maintains references to the grouped files within the block, allowing access to individual files while utilizing shared block space, thereby decreasing storage fragmentation and improving overall storage density.

Inventive Principle:
Principle #5Merging (Combining)

3Device complexity

If each metadata structure is stored in a separate filesystem block, then metadata management is more straightforward, but storage capacity is wasted and the load on the storage system increases

Engineering Contradiction:
Improvemetadata management complexityVSAvoidwasted storage capacity
Core Design Contradiction:
Device complexityVSLoss of substance

Solution Approach 1:

The patent groups multiple metadata structures together in a single filesystem block, reducing the number of blocks required for metadata storage. The filesystem maintains the ability to manage and access individual metadata structures while benefiting from the space efficiency of consolidated storage, thereby reducing wasted capacity and decreasing the load on the underlying storage system.

Inventive Principle:
Principle #5Merging (Combining)

4Reliability

If the filesystem allocates separate blocks for each small object, then object isolation is better and recovery is easier, but storage efficiency decreases and throughput is reduced

Engineering Contradiction:
Improveobject isolation and recoveryVSAvoidstorage throughput
Core Design Contradiction:
ReliabilityVSProductivity

Solution Approach 1:

The patent combines multiple small objects into shared filesystem blocks while maintaining logical separation and individual accessibility. The filesystem structure preserves the ability to isolate and recover individual objects through reference tracking, while the physical consolidation improves storage efficiency and reduces the number of I/O operations required, thereby increasing throughput without sacrificing reliability.

Inventive Principle:
Principle #5Merging (Combining)

Data Source

PatentUS12530318B2Grouping data to conserve storage capacity
Publication Date: 2026.01.20 HITACHI VANTARA LLC
  • US12530318B2 patent drawing
  • US12530318B2 patent drawing
  • US12530318B2 patent drawing

AI summary

In some examples, a computing device stores root metadata data structures (DSs) together in a group root metadata filesystem block, and stores, in a data filesystem block, at least first file data referenced by the first root metadata DS and second file data referenced by a second root metadata DS. The computing device may determine a reference count of the data filesystem block based on the first file data referenced by the first root metadata DS and the second file data referenced by the second root metadata DS. In addition, the computing device may determine a third file data having content that matches the first file data and referenced by a third root metadata DS. The computing device may update the third root metadata DS to reference the first file data, and update the reference count of the data filesystem block based on updating the third root metadata DS.