Storage Array Drive Recommendation Using Workload Telemetry
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for selecting replacement drives for storage arrays are inadequate as they rely on generic MTBF or MTTF ratings, which do not account for the unique workload profiles of individual customers, leading to potential mismatches and increased failure rates.
Innovation Solution
A system that utilizes telemetry data and machine learning to analyze workload characteristics of failed drives, recommending replacement drives tailored to specific workload profiles, considering IO transactions, environmental factors, and customer preferences.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If generic MTBF or MTTF ratings are used to select replacement drives, then the selection process is simple and fast, but the reliability of the storage array decreases due to mismatch between drive capabilities and actual workload requirements
Solution Approach 1:
The patent transforms the drive selection criteria from generic MTBF/MTTF ratings to workload-specific parameters including IOPS, throughput, latency, and other performance metrics that directly reflect actual operational requirements. This parameter transformation enables more accurate matching between drive capabilities and workload demands, thereby improving reliability without excessive complexity
Solution Approach 2:
The system automatically collects telemetry data from the storage array, analyzes workload characteristics, and generates replacement drive recommendations without requiring manual intervention. This self-service approach handles the complexity of workload analysis internally while presenting a simple recommendation output to users, resolving the contradiction between reliability improvement and complexity increase
2Duration of action of moving object
If drive replacement is done without considering workload characteristics, then the replacement process is quick and simple, but the operational life of replacement drives is reduced due to mismatches
Solution Approach 1:
The system performs preliminary analysis of workload characteristics and drive performance requirements before the actual replacement decision is made. By pre-collecting and analyzing telemetry data, the system prepares workload profiles and compatibility assessments in advance, enabling faster replacement decisions that extend drive operational life without incurring time delays during the actual replacement process
Solution Approach 2:
The system continuously monitors drive performance and workload characteristics, using this feedback to refine replacement recommendations. This feedback loop ensures that replacement drives are selected based on actual observed performance data rather than theoretical specifications, extending operational life while minimizing the time needed for selection through learned patterns
3Measurement precision
If telemetry data collection and analysis is implemented, then drive selection accuracy improves, but system complexity and data processing requirements increase
Solution Approach 1:
The patent extracts only the most relevant workload characteristics and performance metrics from the collected telemetry data, rather than analyzing all available data points. This selective extraction focuses computational resources on the most impactful parameters for drive selection, achieving high measurement precision while controlling data processing complexity through targeted analysis
Data Source
AI summary
Architectures and techniques are described that can provide a recommendation for a new drive that is to be added to a storage array. The new drive can be a replacement drive that replaces a failed drive of the storage array or one or more additional drives that expand the storage array. Advantageously, the recommended drive can be specifically tailored to a workload under which the new drive is expected to operate once installed in the storage array. For example, workload metrics or conditions for the failed drive or the storage array can be examined. In response, the recommended drive can be one that is determined to have improved durability (or another characteristic) under the expected service conditions for the new or replacement drive.