Storage I/O Performance Expansion via Dynamic Load Prediction

Solution Overview

Problem

In block storage systems, the input/output (I/O) load often exceeds the upper limit during simultaneous backup and restoration tasks, triggering I/O throttling and potentially causing service interruptions.

Innovation Solution

A method that acquires backup settings and user priority from clients, determines the I/O specification based on these factors, and dynamically adjusts the I/O performance of the storage system to prevent I/O throttling by increasing performance during predicted peak loads.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If the storage system handles simultaneous backup and restoration tasks, then the data protection capability is improved, but the I/O load exceeds the upper limit causing I/O throttling

Engineering Contradiction:

Improvedata protection capabilityVSAvoidI/O throughput

Core Design Contradiction:

ReliabilityVSProductivity

Solution Approach 1:

The system performs preliminary actions by predicting peak I/O load times based on backup settings and user priority before they occur. The storage system proactively increases I/O performance in advance of the predicted peak load, preventing I/O throttling before it happens rather than reacting after the problem occurs.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The storage system dynamically adjusts its I/O performance characteristics based on real-time conditions. By monitoring current I/O load and comparing it with predicted peak values, the system dynamically increases or maintains I/O performance during critical periods and adjusts accordingly, transforming a static storage system into one that adapts to varying workloads.

Inventive Principle:

Principle #15Dynamics

2Reliability

If the storage system increases I/O performance to handle peak loads, then service interruptions are prevented, but resource consumption increases

Engineering Contradiction:

Improveservice continuityVSAvoidresource consumption

Core Design Contradiction:

ReliabilityVSUse of energy by moving object

Solution Approach 1:

The system performs preliminary actions by predicting peak I/O load times based on backup settings and user priority before they occur. The storage system proactively increases I/O performance in advance of the predicted peak load, preventing I/O throttling before it happens rather than reacting after the problem occurs.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The system changes key performance parameters (I/O throughput, latency) dynamically based on predicted workload conditions. By adjusting these parameters in advance during low-demand periods and maintaining them during predicted peak periods, the system optimizes resource consumption while ensuring service continuity.

Inventive Principle:

Principle #35Parameter changes

3Productivity

If the storage system allocates higher I/O specification to users, then user performance needs are met, but the system capacity for other users is reduced

Engineering Contradiction:

Improveuser I/O performanceVSAvoidsystem capacity flexibility

Core Design Contradiction:

ProductivityVSAdaptability or versatility

Solution Approach 1:

The system applies local quality by assigning different I/O performance characteristics to different users or user groups based on their priority levels and specific needs. High-priority users receive enhanced I/O performance during peak periods, while lower-priority users receive appropriate service levels, creating a differentiated service model that optimizes overall system utilization.

Inventive Principle:

Principle #3Local quality

Solution Approach 2:

The storage system dynamically adjusts its I/O performance characteristics based on real-time conditions. By monitoring current I/O load and comparing it with predicted peak values, the system dynamically increases or maintains I/O performance during critical periods and adjusts accordingly, transforming a static storage system into one that adapts to varying workloads.

Inventive Principle:

Principle #15Dynamics

Data Source

PatentUS12277035B2Method, electronic device, and computer program product for storage performance expansion

Publication Date: 2025.04.15 DELL PROD LP

AI summary

Embodiments of the present disclosure provide a method, an electronic device, and a computer program product for storage performance expansion. The method includes acquiring, from a client, backup settings of a user for backing up data in a storage system and the user's priority. The method further includes determining, based on the user's priority, an input/output (I/O) specification of the storage system that is able to be allocated to the client. The method further includes determining, based on the backup settings, a time period in which a peak value of an I/O load of the storage system occurs. The method further includes increasing I/O performance of the storage system in response to that the peak value is greater than an upper limit of the I/O specification. The embodiments of the present disclosure can solve the I/O peak problem more effectively without interrupting a data backup or restoration service.