Storage System Data Reordering for Random Access Latency

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing storage systems face high random access latency, leading to suboptimal access times due to data segments being stored in non-sequential orders, which conventional defragmentation techniques only partially address.

Innovation Solution

A learning or content-based analysis algorithm identifies data access patterns and reorders data segments to facilitate sequential access, rearranging them to be stored contiguously based on predicted access sequences, thereby improving access efficiency.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If data segments are stored in contiguous blocks on disk, then reading efficiency is improved, but data segments must be accessed in storage order rather than access pattern order

Engineering Contradiction:
Improvereading efficiencyVSAvoidaccess flexibility
Core Design Contradiction:
ProductivityVSEase of operation

Solution Approach 1:

The system performs preliminary actions by identifying access patterns and pre-reordering data segments according to those patterns before actual access occurs. The learning algorithm analyzes historical access patterns and proactively reorganizes data segments to match predicted future access sequences, so that when access occurs, the data is already in optimal position for sequential reading.

Inventive Principle:
Principle #10Preliminary action

2Loss of time

If conventional defragmentation is used to improve access time, then some access performance is improved, but only limited improvements are achieved because it does not address random access patterns

Engineering Contradiction:
Improveaccess timeVSAvoidimprovement effectiveness
Core Design Contradiction:
Loss of timeVSProductivity

Solution Approach 1:

The system implements feedback by using a learning algorithm that continuously monitors and analyzes actual data access patterns. The algorithm processes access history information and uses this feedback to dynamically adjust and refine predictions of future access patterns, which then guide the reordering of data segments. This closed-loop approach enables the system to adapt to changing access behaviors and continuously optimize access time.

Inventive Principle:
Principle #23Feedback

Solution Approach 2:

The system changes the parameter of data segment arrangement from static contiguous storage to dynamic pattern-based ordering. By transforming the organization parameter according to learned access patterns, the system converts random access operations into sequential access operations, dramatically reducing access time and improving effectiveness beyond conventional defragmentation.

Inventive Principle:
Principle #35Parameter changes

3Ease of operation

If disk head moves to distant data blocks to access scattered data segments, then all data segments can be accessed, but access time increases due to frequent head movement

Engineering Contradiction:
Improvedata accessibilityVSAvoidaccess time
Core Design Contradiction:
Ease of operationVSLoss of time

Solution Approach 1:

The system performs preliminary reordering of data segments based on predicted access patterns before access requests arrive. By proactively organizing data segments in the sequence they are likely to be accessed, the system eliminates the need for frequent disk head movements during actual access operations, converting random access into sequential access and significantly reducing access time while maintaining full data accessibility.

Inventive Principle:
Principle #10Preliminary action

Data Source

PatentUS9250819B2Learning machine to optimize random access in a storage system
Publication Date: 2016.02.02 QUEST SOFTWARE INC
  • US9250819B2 patent drawing
  • US9250819B2 patent drawing
  • US9250819B2 patent drawing

AI summary

Mechanisms are provided for optimizing random access in a storage system. According to various embodiments, an access pattern may be identified for a plurality of data segments stored in a first arrangement on a storage medium. Each of the plurality of data segments may be stored at a respective first storage location on the storage medium in the first arrangement. The access pattern may indicate an order in which the data segments are likely to be retrieved from the storage medium. The plurality of data segments may be stored in a second arrangement on the storage medium based on the identified access pattern. Each of the plurality of data segments may be stored at a respective second storage location on the storage medium in the updated arrangement.