Storage Device Wear and Health State Prediction Dataset

#health state prediction #remaining useful life estimation #time series forecasting #HDD and SSD health monitoring #storage device failure prediction #predictive maintenance

5000
0.3GB
JSON
CATL
MOBIUSI INC

Updated:2026-11-04

AI Analysis & Value Prop

This simulated time-series dataset represents HDD and SSD workloads, device telemetry, and lifecycle changes, with observations such as read and write volume, IOPS, temperature, media wear, and cumulative power-on hours. It also includes targets and estimates such as current and future health status, degradation risk, remaining useful life, and predicted wear, making it suitable for studying how storage hardware health changes over time. Use it to train and evaluate models for health prediction, wear trend analysis, failure warning, and remaining-life estimation in storage operations and predictive maintenance research.

Dataset Insights

Sample Examples

{}

Technical Specifications

Field	Type	Description
file_name	string	File name
read_iops	number	Number of read operations per second during the observation interval.
write_iops	number	Number of write operations per second during the observation interval.
device_type	string	Type of storage device, such as HDD or SSD.
sample_time	string	Timestamp associated with this device observation.
power_on_hours	number	Total time the device has been powered on up to the observation time, in hours.
media_wear_percent	number	Media wear level recorded by device telemetry, expressed as a percentage.
temperature_celsius	number	Device temperature recorded by telemetry, in degrees Celsius.
workload_read_bytes	number	Amount of data read from the device during the observation interval, in bytes.
future_health_status	string	Health status at a future target time, derived from current and historical observations or provided as a prediction target.
workload_write_bytes	number	Amount of data written to the device during the observation interval, in bytes.
current_health_status	string	Device health status at the observation time, such as healthy, warning, or failed.
predicted_wear_percent	number	Predicted media wear level at the end of the prediction horizon, expressed as a percentage.
health_degradation_risk	number	Risk score for device health degradation within the prediction horizon, ranging from 0 to 1.
estimated_remaining_life_hours	number	Estimated remaining usable device life based on its lifecycle state, in hours.

Compliance Statement

Authorization Type	Proprietary - Commercial AI Training License (No Redistribution)
Commercial Use	Requires exclusive subscription or authorization contract (monthly or per-invocation charging)
Privacy and Anonymization	No PII, no real company names, simulated scenarios follow industry standards
Compliance System	Compliant with China's Data Security Law / EU GDPR / supports enterprise data access logs

Frequently Asked Questions

Which devices and telemetry indicators are included?: The dataset covers HDD and SSD scenarios and includes time-series indicators such as read and write volume, read and write IOPS, device temperature, media wear, and cumulative power-on hours. It also provides current and future health status, degradation risk, estimated remaining useful life, and predicted wear.

Are these records collected from real devices?: This is simulated data designed to represent relationships among storage workloads, telemetry, and health changes. It is suitable for model development, method validation, and educational research, but should not be treated as measured records from real hardware or as a guarantee of production performance.

Which model training and evaluation tasks does it support?: Potential tasks include health state classification or prediction, time-series forecasting, media wear trend analysis, health degradation risk assessment, and remaining useful life estimation. You can build historical windows from the observation time and feature fields, then split the data according to the intended prediction target.

How can I obtain the data, and what are the license terms?: Availability, delivery formats, and license scope should be confirmed on the product page or in the delivery documentation. Before use, check whether the license covers your intended research, commercial deployment, or redistribution; the supplied information does not specify a download method or license terms.

Can't find the data you need?

Post a request and let data providers reach out to you.

Get this Dataset

Verified for Enterprise Use

Cite this Work

@dataset{Mobiusi2026,
  title={Storage Device Wear and Health State Prediction Dataset},
  author={MOBIUSI INC},
  year={2026},
  url={https://www.mobiusi.com/datasets/ba93fdf86ccebb63c9ac5cf06f312590?dataset_scene_id=6},
  urldate={2026-10-20},
  keywords={storage device health prediction,HDD life prediction,SSD wear prediction,drive failure warning,storage telemetry data,time-series forecasting,remaining life estimation,device health status},
  version={1.0}
}

Using this in research? Please cite us.

placeholder

Scene

Domain

Modality

Task