Dynamic Cluster Allocation for Storage Write Performance

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current information processing devices, such as personal computers, smartphones, and tablets, statically apply a single cluster allocation policy for various data writing methods, leading to reduced write performance, shorter storage medium lifespan, and file system fragmentation.

Innovation Solution

An information processing apparatus and method that dynamically determines the allocation policy for clusters based on the number of files, file size, data type, and write requests, using a controller to select blocks for writing, optimizing between sequential and parallel allocation policies for improved performance.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Device complexity

If a single static cluster allocation policy is used for all data writing methods, then the device complexity is reduced, but the write performance is lowered and storage medium lifespan is shortened

Engineering Contradiction:
Improveallocation policy complexityVSAvoidwrite performance
Core Design Contradiction:
Device complexityVSProductivity

Solution Approach 1:

The patent implements dynamic cluster allocation policies that automatically adapt based on the type of data being written. The system monitors write operations and switches between different allocation strategies (sequential allocation for continuous writes, parallel allocation for multiple simultaneous writes) to optimize performance for each specific writing scenario rather than using a fixed single policy.

Inventive Principle:
Principle #15Dynamics

Solution Approach 2:

The patent changes the allocation policy parameters dynamically based on detected writing conditions. When continuous writing is detected, the system switches to sequential allocation with specific cluster size parameters; when multiple simultaneous writes are detected, it switches to parallel allocation with different cluster selection parameters, thereby optimizing performance for each writing pattern.

Inventive Principle:
Principle #35Parameter changes

2Ease of operation

If a single static cluster allocation policy is used for all data writing methods, then the ease of operation is maintained, but the storage medium lifespan is shortened

Engineering Contradiction:
Improveoperation simplicityVSAvoidstorage medium lifespan
Core Design Contradiction:
Ease of operationVSDuration of action of stationary object

Solution Approach 1:

The system performs self-optimization by automatically detecting the type of write operation and selecting the appropriate allocation policy without user intervention. The controller monitors write patterns, identifies whether continuous or parallel writing occurs, and autonomously switches allocation strategies to protect storage medium lifespan while maintaining transparent operation for the user.

Inventive Principle:
Principle #25Self-service

3Device complexity

If a single static cluster allocation policy is used for all data writing methods, then the device complexity is reduced, but file system fragmentation occurs

Engineering Contradiction:
Improveallocation policy complexityVSAvoidfile system fragmentation
Core Design Contradiction:
Device complexityVSStability of the object's composition

Solution Approach 1:

The patent segments the allocation policy into distinct strategies tailored for different writing scenarios. It separates sequential allocation for continuous writes from parallel allocation for multiple simultaneous writes, allowing each segment to manage cluster assignment independently and prevent fragmentation by keeping related data blocks contiguous within their respective allocation contexts.

Inventive Principle:
Principle #1Segmentation

Data Source

PatentUS9904620B2Information processing apparatus and method for writing data to a storage medium
Publication Date: 2018.02.27 SONY GROUP CORP
  • US9904620B2 patent drawing
  • US9904620B2 patent drawing
  • US9904620B2 patent drawing

AI summary

An information processing apparatus includes a controller configured to obtain the number of files to be written to a storage medium whose storage area configured in units of a predetermined block and whose stored data is erased in units of the predetermined block, and to determine a policy for selecting, based on the obtained number of files, the block to which the one or more files are to be written.