Overview NVMe Fleet Analytics
Total Drives
Fleet inventory
Healthy
Failed
SMART Flagged
Warning active
High Temp
>55°C drives
High Life Used
>80% endurance
Failure Mode Breakdown
Distribution of failed drives by cause
Failure Rate by Vendor
% failed drives per vendor
Temperature Distribution
Drive count by thermal bucket
Life Used Distribution
Flash endurance consumed
Power-On Hours vs Life Used — Failure Scatter
Cyan = healthy sample · Red = failed drives
Vendor Health Matrix
VendorTotalFailedFail RateAvg TempAvg Life
Model Performance
ModelTotalFailedFail RateAvg Life
Firmware Failure Rates
Model Failure Rate Comparison
Error Metrics by Vendor — Radar
Mean error rates across all drives
Read vs Write Error Rate
Media & CRC Errors by Vendor
Active Alerts — SMART Flagged Drives
Drives with SMART warning active, sorted by life used
Drive IDVendorModelFirmware POH (hrs)TBWTemp °C Life %Media ErrCRC Err Read ERRSMARTStatusFailure Mode
Model Configuration
Algorithm
Random Forest Classifier
Feature Importances
Contribution to prediction accuracy
REST API Endpoints
Backend endpoints served by server.py
MethodEndpointDescription
GET/api/fleet/summaryFleet KPI summary
GET/api/fleet/failure_modesFailure mode distribution
GET/api/fleet/by_vendorStats grouped by vendor
GET/api/fleet/by_modelStats grouped by model
GET/api/fleet/by_firmwareStats grouped by firmware
GET/api/fleet/temperature_distTemperature distribution buckets
GET/api/fleet/life_distLife-used distribution buckets
GET/api/fleet/scatterScatter plot data (POH vs life)
GET/api/fleet/error_by_vendorError metrics by vendor
GET/api/drivesPaginated drive list (filter/sort)
GET/api/alertsSMART-flagged alert drives
POST/api/predictSingle drive prediction
POST/api/predict/batchBatch prediction (up to 500)
GET/api/model/infoModel metadata & feature importances