Storage Device Survival Prediction Using Conformal Prediction Framework
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing predictive technologies for storage devices, such as Dell EMC's SupportAssist, do not account for the potential rapid deterioration of storage devices between the initial issue detection and remedial action, leading to unexpected failures and productivity losses.
Innovation Solution
A conformal prediction framework combined with an online semi-parametric Mondrian survival forest classifier is used to predict the survival rate of storage devices in real-time, providing frequent updates on their health and enabling timely remedial actions based on recent telemetry data.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If predictive technology detects storage device issues and creates support cases, then customer support response is initiated, but the storage device condition may worsen or deteriorate rapidly before remedial action is taken
Solution Approach 1:
The system performs preliminary actions by continuously monitoring storage device health parameters and predicting potential failures before they occur. The conformal prediction framework provides advance warning with confidence intervals, allowing proactive remedial actions to be taken before the actual failure event, thus preventing the worsening of device condition during the response window.
Solution Approach 2:
The system implements continuous feedback loops where storage device telemetry data is constantly monitored, processed through the conformal prediction framework, and used to update risk assessments in real-time. This feedback mechanism allows the system to detect deterioration trends and adjust predictions dynamically, providing ongoing guidance for timely remedial actions.
2Measurement precision
If storage device health is monitored continuously, then real-time survival rate prediction is possible, but system complexity increases
Solution Approach 1:
The conformal prediction framework serves multiple functions simultaneously: it provides survival rate predictions, calculates confidence intervals, quantifies uncertainty, and handles censored data. This multi-functionality reduces the need for separate specialized components, thereby managing system complexity while maintaining high measurement precision for real-time health monitoring.
Solution Approach 2:
The system manages complexity by transforming diverse storage device parameters into a unified survival time prediction framework. The conformal prediction framework standardizes the output to confidence intervals and survival rates, making it easier to interpret and act on regardless of the specific underlying hardware parameters being monitored.
Data Source
AI summary
The described technology is generally directed towards predicting the survival of a storage device (e.g., a hard disk drive or a solid state drive) to a specified time point, expressed as a confidence score, via a prediction uncertainty quantifier framework with a machine learning classifier. The confidence score corresponds to the likelihood of a storage device surviving until a specified time point (e.g., for n hours). In one implementation, a conformal prediction framework provides the confidence score for a storage device, based on survival rate data predicted using recent telemetry data collected for that storage device by an online semi-parametric Mondrian survival forest classifier. Updated confidence scores based on updated telemetry data can be obtained at various evaluation stages to reevaluate whether to take remedial action with respect to a storage device (e.g., replace the storage device). Multiple storage devices can be ranked by their respective associated confidence scores.