Automated Storage Capacity Provisioning Using Machine Learning
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional storage management approaches face challenges in determining appropriate storage capacity provisioning in response to alerts, often leading to errors and inefficient resource allocation due to reliance on existing knowledge and thin provisioning imperfections.
Innovation Solution
The implementation of automated storage capacity provisioning using machine learning techniques, which processes user-provided inputs and historical data to determine the duration for which provisioned storage capacity will last, enabling more accurate and efficient storage planning.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional storage management approaches are used, then storage capacity expansion is performed based on alerts, but determination of appropriate storage capacity amounts is difficult and error-prone
Solution Approach 1:
The system performs self-service by automatically determining appropriate storage capacity amounts using machine learning techniques. The storage management system analyzes historical data and current alerts to autonomously calculate required capacity expansions, eliminating reliance on administrator knowledge and reducing human error in capacity planning decisions.
Solution Approach 2:
The patent replaces manual mechanical decision-making processes with automated machine learning algorithms. Instead of administrators manually analyzing alerts and determining capacity needs based on experience, the system uses ML models to automatically process data and generate accurate capacity recommendations, substituting human cognitive processes with computational algorithms.
2Adaptability or versatility
If thin provisioning is used, then storage allocation flexibility is improved, but capacity planning errors and imperfections are overlooked
Solution Approach 1:
The system implements feedback mechanisms that continuously monitor storage utilization and compare actual usage against planned allocations. By analyzing historical data and current system state, the ML models provide feedback on capacity planning accuracy, identifying errors and imperfections that thin provisioning might otherwise hide, and enabling corrective actions to improve future planning reliability.
3Measurement precision
If automated machine learning techniques are used, then storage capacity determination accuracy is improved, but system complexity increases
Solution Approach 1:
The patent applies universality by designing a multi-functional storage management system that integrates multiple capabilities: data collection, historical analysis, machine learning model execution, alert processing, and capacity recommendation generation. This universal system handles various storage scenarios and data types through a single unified platform, managing complexity through functional integration rather than separate specialized components.
Data Source
AI summary
Methods, apparatus, and processor-readable storage media for automated storage capacity provisioning using machine learning techniques are provided herein. An example computer-implemented method includes obtaining a user-provided input comprising an identification of an amount of storage capacity to be provisioned from a storage system; determining an amount of time for which the amount of storage capacity to be provisioned will last in connection with the storage system by processing the user-provided input in connection with historical data pertaining to storage utilization using one or more machine learning techniques; outputting, to the user, the determined amount of time for which the amount of storage capacity to be provisioned will last; and performing one or more automated actions based at least in part on feedback from the user in response to the outputting of the determined amount of time for which the amount of storage capacity to be provisioned will last.