File System Delete Operations Bulk Transaction Management

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Conventional file deletion operations in file systems are inefficient, particularly for large files, as they require traversing the entire file system hierarchy, generating a high number of metadata transactions and I/O requests, which consumes significant system resources and time.

Innovation Solution

The method involves creating bulk sub-transactions that store multiple metadata transactions for each leaf indirect data block, updating contents only when data blocks are freed, and using multiple threads to process leaf indirect blocks concurrently, reducing the number of metadata transactions and I/O operations.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If conventional file deletion operations are used, then data integrity is maintained, but system resources and time are consumed significantly

Engineering Contradiction:
Improvedata integrityVSAvoiddeletion time
Core Design Contradiction:
ReliabilityVSLoss of time

Solution Approach 1:

The patent merges multiple metadata transactions into a single bulk transaction. Instead of processing each data block deletion separately and generating individual metadata transactions for each, the system combines all deletions within a leaf indirect block into one atomic transaction, significantly reducing the number of I/O operations and time required while maintaining data integrity through proper transaction management

Inventive Principle:
Principle #5Merging (Combining)

Solution Approach 2:

The patent segments the file system hierarchy into leaf indirect blocks as independent processing units. Each leaf indirect block can be processed and deleted independently from other blocks, enabling parallel processing and reducing the overall deletion time while maintaining integrity through proper transaction boundaries and parent block updates

Inventive Principle:
Principle #1Segmentation

2Reliability

If conventional file deletion operations are used, then data integrity is maintained, but the number of metadata transactions and I/O requests increases

Engineering Contradiction:
Improvedata integrityVSAvoidnumber of metadata transactions
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The patent merges multiple individual metadata transactions into a single bulk metadata transaction. When data blocks within a leaf indirect block are deleted, instead of generating separate metadata transactions for each block, the system creates one bulk transaction that records all deletions, thereby reducing the number of I/O requests and system complexity while maintaining data integrity through proper transaction logging

Inventive Principle:
Principle #5Merging (Combining)

Solution Approach 2:

The patent performs preliminary actions by updating the parent indirect block's free block list before actually deleting the leaf indirect block. This preliminary update prepares the system for efficient deletion by ensuring the parent block is already aware of the upcoming deletion, reducing the need for subsequent I/O operations and metadata transactions

Inventive Principle:
Principle #10Preliminary action

3Device complexity

If single-threaded processing is used, then process simplicity is maintained, but processing speed is reduced

Engineering Contradiction:
Improveprocess simplicityVSAvoiddeletion speed
Core Design Contradiction:
Device complexityVSProductivity

Solution Approach 1:

The patent segments the file deletion process into independent units (leaf indirect blocks) that can be processed concurrently by multiple threads. Each thread handles a specific leaf indirect block independently, allowing parallel processing that significantly increases deletion speed while maintaining relative simplicity through proper synchronization and block-level independence

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent introduces dynamic multi-threaded processing that adapts to the file system structure. The system dynamically creates and manages threads based on the number and size of leaf indirect blocks, allowing the processing model to adjust between single-threaded (for simplicity) and multi-threaded (for performance) modes depending on the specific deletion scenario

Inventive Principle:
Principle #15Dynamics

Data Source

PatentUS9442955B1Managing delete operations in files of file systems
Publication Date: 2016.09.13 EMC IP HLDG CO LLC
  • US9442955B1 patent drawing
  • US9442955B1 patent drawing
  • US9442955B1 patent drawing

AI summary

A method is used in managing delete operations in files of file systems. An indirect data block of a file of a file system is deleted. A parent indirect data block pointing to the indirect data block is removed from a list. The parent indirect data block has been added to the list for flushing contents of the parent indirect data block to a storage device.