Storage Workload Classification Using ML Time-Series Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional storage management techniques fail to provide sufficient insight into the input/output characteristics of workloads, leading to reduced performance and resource wastage in storage systems due to inadequate workload understanding.
Innovation Solution
The use of machine learning techniques to automatically detect workload types by designating performance-related characteristics, generating supervised datasets, and training models to analyze historical time series data for workload classification and subsequent automated actions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional storage management techniques are used, then device complexity is reduced, but measurement precision of workload characteristics deteriorates
Solution Approach 1:
The patent introduces machine learning models as intermediary components between storage systems and management interfaces. These models automatically analyze workload characteristics and translate raw performance data into actionable insights, thereby improving measurement precision without requiring complex manual management systems.
Solution Approach 2:
The system implements self-service through automated workload classification and characterization using machine learning. The storage system automatically detects and categorizes workload types without human intervention, providing precise workload insights while maintaining simple management operations.
2Productivity
If multiple storage systems are deployed to distribute workloads, then productivity is improved, but loss of substance increases due to resource wastage
Solution Approach 1:
The patent implements feedback mechanisms where machine learning models continuously monitor storage system performance and workload characteristics. Based on this feedback, the system dynamically optimizes resource allocation and workload distribution, improving productivity while minimizing resource wastage by avoiding unnecessary expansion to multiple storage systems.
Solution Approach 2:
The system changes operational parameters through automated workload classification. By identifying workload types and their characteristics, the system adjusts storage resource allocation parameters dynamically, enabling efficient utilization of existing resources and preventing resource wastage associated with over-provisioning or premature system expansion.
3Adaptability or versatility
If storage systems overbuy capacity to support multiple workloads, then adaptability is improved, but loss of energy increases
Solution Approach 1:
The patent applies dynamics by implementing adaptive workload classification and resource allocation. The machine learning models continuously learn from incoming data and dynamically adjust storage resource allocation based on actual workload characteristics and patterns, enabling the system to adapt to multiple workload types without over-provisioning and wasting energy.
Data Source
AI summary
Methods, apparatus, and processor-readable storage media for automatically detecting workload type-related information in storage systems using machine learning techniques are provided herein. An example computer-implemented method includes designating workload types associated with a storage system by identifying performance-related characteristics associated with each of the workload types; obtaining historical time series data, associated with the storage system, pertaining to multiple performance-related metrics; partitioning the historical time series data into multiple segments based on value levels of the performance-related metrics; generating supervised datasets by associating, for each of the segments, the value levels of the performance-related metrics with a given one of the designated workload types; training a machine learning model using the supervised datasets; detecting workload types associated with input data from the storage system by processing the input data using the machine learning model; and performing automated actions based on the detected workload types.