Atomic Bitmap Bit Clearing for Concurrent Disk Access

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current disk command sets lack the ability to change the state of a single bit in a bitmap on disk without reading and writing the entire structure, leading to potential corruption when multiple computers attempt to update different bits concurrently, and existing solutions like exclusive access create bottlenecks or single points of failure.

Innovation Solution

A method and apparatus that utilize a BitmapSetClear command to atomically clear one or more bits in a bitmap on disk storage, incorporating synchronization data to distinguish between current and previous generations, allowing concurrent access without exclusive access requirements.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If exclusive access is implemented to prevent bitmap corruption during updates, then reliability is improved, but productivity deteriorates due to bottlenecks and single points of failure

Engineering Contradiction:
Improvebitmap data integrityVSAvoiddistributed application operation
Core Design Contradiction:
ReliabilityVSProductivity

Solution Approach 1:

The bitmap update operation is segmented into atomic read-modify-write steps that are executed independently for each bit or group of bits. Each initiator can perform its own segmented updates without requiring exclusive access to the entire bitmap, thereby maintaining reliability while improving productivity through concurrent access.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The system changes the parameter of access mode from exclusive to concurrent by implementing version vectors and conflict detection mechanisms. This allows multiple initiators to access the bitmap simultaneously with different parameters (version identifiers, bit positions), resolving the contradiction between reliability and productivity.

Inventive Principle:
Principle #35Parameter changes

2Manufacturing precision

If read-update-write cycle is implemented for single bit changes on disk, then manufacturing precision is improved, but loss of time increases due to reading and writing entire bitmap structure

Engineering Contradiction:
Improvebitmap update accuracyVSAvoidbitmap update duration
Core Design Contradiction:
Manufacturing precisionVSLoss of time

Solution Approach 1:

The patent extracts only the necessary portion of the bitmap (specific bits to be updated) rather than reading and writing the entire bitmap structure. This is achieved through selective bit manipulation commands that operate on specific bit positions, thereby maintaining update accuracy while significantly reducing the time required for bitmap updates.

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

Instead of performing a complete read-update-write cycle on the entire bitmap, the system performs partial action by updating only the specific bits that need to be changed. This partial action approach maintains the precision of the update while minimizing the time loss associated with transferring unnecessary data.

Inventive Principle:
Principle #16Partial or excessive action

3Reliability

If global locks or designated bitmap manager are used to establish exclusive access, then reliability is improved, but device complexity increases

Engineering Contradiction:
Improveconcurrent access controlVSAvoidapplication architecture
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

Each initiator performs its own bitmap update operations independently without requiring a designated bitmap manager or global locks. The system provides self-service mechanisms through atomic bit manipulation commands and version vector validation, allowing each initiator to manage its own access control and reducing overall system complexity while maintaining reliability.

Inventive Principle:
Principle #25Self-service

Data Source

PatentUS8065489B1Method and apparatus for managing concurrent access among computers to a bitmap stored on disk storage
Publication Date: 2011.11.22 ARCTERA US LLC
  • US8065489B1 patent drawing
  • US8065489B1 patent drawing
  • US8065489B1 patent drawing

AI summary

Method and apparatus for managing concurrent access among computers to a bitmap stored on disk storage. In some examples, a command is received from a requesting computer of the computers, the command indicating that one or more bits in the bitmap are to be cleared. The bitmap as stored on the disk storage is updated responsive to the command to clear the one or more bits in the bitmap. Synchronization data associated with the bitmap is stored. The synchronization data is configured to distinguish between a current generation and a previous generation of the bitmap, and to specify an action to be performed in response to receiving a subsequent command for accessing the bitmap if the subsequent command identifies a previous generation.