Dynamic Backup Stream Configuration for Throughput Optimization

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing backup systems face challenges in efficiently managing increasing data volumes, leading to prolonged backup windows, as simply splitting data into multiple streams can adversely impact performance due to poor fit with data and hardware characteristics.

Innovation Solution

Dynamic backup stream configuration, assessing data characteristics such as read block size and concurrency factors to optimize net aggregate throughput, splitting data into backup streams that account for specific hardware and data characteristics, and adjusting the number of concurrent streams to maximize throughput.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If data is split into multiple backup streams, then backup throughput may increase, but backup performance deteriorates due to poor fit with data and hardware characteristics

Engineering Contradiction:

Improvebackup throughputVSAvoidbackup performance

Core Design Contradiction:

ProductivityVSReliability

Solution Approach 1:

The patent implements dynamic backup streams that automatically adjust the number of concurrent streams and I/O block sizes based on real-time assessment of data characteristics and hardware performance. This dynamic adaptation resolves the contradiction by optimizing throughput while maintaining compatibility with specific data and hardware configurations, preventing performance deterioration.

Inventive Principle:

Principle #15Dynamics

Solution Approach 2:

The system changes key parameters such as the number of concurrent backup streams and I/O block sizes based on assessed data characteristics and hardware capabilities. By dynamically adjusting these parameters rather than using fixed configurations, the system achieves high throughput while maintaining optimal performance for different data types and hardware configurations.

Inventive Principle:

Principle #35Parameter changes

2Speed

If the number of concurrent backup streams is increased, then data transfer rate may improve, but backup window duration increases due to system overhead

Engineering Contradiction:

Improvedata transfer rateVSAvoidbackup window duration

Core Design Contradiction:

SpeedVSDuration of action of moving object

Solution Approach 1:

The patent dynamically determines the optimal number of concurrent streams during the backup operation based on assessed hardware capabilities and data characteristics. This prevents premature commitment to suboptimal stream counts, ensuring high data transfer rates are achieved without extending the backup window due to excessive system overhead from too many streams.

Inventive Principle:

Principle #15Dynamics

3Device complexity

If static backup stream configuration is used, then system complexity is reduced, but adaptability to different data and hardware characteristics deteriorates

Engineering Contradiction:

Improvebackup system complexityVSAvoidadaptability to data and hardware

Core Design Contradiction:

Device complexityVSAdaptability or versatility

Solution Approach 1:

The backup system performs self-assessment of data characteristics and hardware capabilities, then automatically configures optimal backup streams without requiring complex manual configuration. This self-service approach maintains low system complexity while achieving high adaptability to different data and hardware configurations through automated parameter optimization.

Inventive Principle:

Principle #25Self-service

Data Source

PatentUS9658926B1Systems and methods for dynamic save streams

Publication Date: 2017.05.23 EMC IP HLDG CO LLC

AI summary

Systems and methods are provided for performing a backup operation in a computing system. The source data and associated hardware are assessed to determine an optimum read block size and an optimum number of backup streams. The source data is then backed up using the optimum number of backup streams at the optimum read block size.