marshalf/nas-drive-data
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

Refresh dataset: add acoustics + helium fields (2026-06-17 snapshot)

Browse source
This commit is contained in:
deeddy 2026-06-17 06:36:17 +02:00
parent 06bccb4ea7
commit 65ee575668
8 changed files with 300 additions and 290 deletions
Download patch file Download diff file Expand all files Collapse all files

26
README.md
View file

@ -10,11 +10,11 @@ Maintained at [nasdisks.com](https://www.nasdisks.com/), where the same data pow
| File | Rows | What it is |
|------|------|------------|
| [`data/drives-2026-06-16.csv`](data/drives-2026-06-16.csv) | 140+ | The headline file: one flat row per drive, specs joined to failure rate. |
| [`data/drives-2026-06-16.json`](data/drives-2026-06-16.json) | 140+ | The same rows as JSON (mirror of the live API). |
| [`data/cmr-reference-2026-06-16.csv`](data/cmr-reference-2026-06-16.csv) | 140+ | Source of truth for specs + CMR/SMR classification, with a per-row `confidence` flag. |
| [`data/backblaze-afr-2026-06-16.csv`](data/backblaze-afr-2026-06-16.csv) | 36 | Full-year 2025 Backblaze failure rates per model (the models Backblaze runs at scale). |
| [`data/series-reliability-2026-06-16.csv`](data/series-reliability-2026-06-16.csv) | - | Manufacturer series specs (warranty, rated workload, MTBF) used as context where Backblaze has no data. |
| [`data/drives-2026-06-17.csv`](data/drives-2026-06-17.csv) | 140+ | The headline file: one flat row per drive, specs joined to failure rate. |
| [`data/drives-2026-06-17.json`](data/drives-2026-06-17.json) | 140+ | The same rows as JSON (mirror of the live API). |
| [`data/cmr-reference-2026-06-17.csv`](data/cmr-reference-2026-06-17.csv) | 140+ | Source of truth for specs + CMR/SMR classification (now including idle/seek noise and helium fill), with a per-row `confidence` flag. |
| [`data/backblaze-afr-2026-06-17.csv`](data/backblaze-afr-2026-06-17.csv) | 36 | Full-year 2025 Backblaze failure rates per model (the models Backblaze runs at scale). |
| [`data/series-reliability-2026-06-17.csv`](data/series-reliability-2026-06-17.csv) | - | Manufacturer series specs (warranty, rated workload, MTBF) used as context where Backblaze has no data. |
## Quick start
@ -28,7 +28,7 @@ curl -s https://www.nasdisks.com/data/drives.json | jq '.drives[0]'
curl -s https://www.nasdisks.com/data/drives.csv
```
## Columns (`drives-2026-06-16.csv` / `drives-2026-06-16.json`)
## Columns (`drives-2026-06-17.csv` / `drives-2026-06-17.json`)
| Column | Meaning |
|--------|---------|
@ -41,6 +41,8 @@ curl -s https://www.nasdisks.com/data/drives.csv
| `interface` | SATA or SAS. |
| `form_factor` | 3.5 or 2.5. |
| `recording_tech` | `cmr` or `smr` - the field that matters most for RAID/NAS use. |
| `acoustic_idle_db` / `acoustic_seek_db` | Idle / seek noise as A-weighted sound power (dB), from manufacturer datasheets. Null where the maker doesn't publish it. |
| `is_helium` | Whether the drive is helium-sealed (`true`/`false`); null if unconfirmed. Helium is standard on high-capacity enterprise/NAS drives. |
| `drive_class` | NAS, NAS-Pro, Enterprise, Surveillance, Desktop, etc. |
| `in_production` | Whether the drive is currently sold new. |
| `afr_pct` | Backblaze annualized failure rate, full-year 2025 (null where Backblaze doesn't track the model). |
@ -49,9 +51,10 @@ curl -s https://www.nasdisks.com/data/drives.csv
## Methodology
- **CMR vs SMR** is classified from manufacturer datasheets and model-number decoding. SMR drives shingle their tracks, which cripples random-write and RAID-rebuild performance, so the distinction is the single most important spec for a NAS - and manufacturers do not always advertise it. The `confidence` column in `cmr-reference-2026-06-16.csv` flags how certain each call is.
- **CMR vs SMR** is classified from manufacturer datasheets and model-number decoding. SMR drives shingle their tracks, which cripples random-write and RAID-rebuild performance, so the distinction is the single most important spec for a NAS - and manufacturers do not always advertise it. The `confidence` column in `cmr-reference-2026-06-17.csv` flags how certain each call is.
- **Noise and helium** are read from manufacturer datasheets (idle/seek A-weighted sound power; helium-seal designation) and are licensed CC BY 4.0 alongside the other specs.
- **Failure rates** are computed from Backblaze's raw 2025 quarterly Drive Stats, aggregated to a full-year annualized rate and validated against Backblaze's own published fleet figure (~1.36% for 2025). Only the catalog models that exactly match a Backblaze-tracked model carry an `afr_pct`; the rest are left null rather than guessed.
- **`series-reliability-2026-06-16.csv`** gives manufacturer-rated context (warranty years, workload TB/year, MTBF) for lines Backblaze doesn't run, so you have *something* to reason about beyond a blank cell.
- **`series-reliability-2026-06-17.csv`** gives manufacturer-rated context (warranty years, workload TB/year, MTBF) for lines Backblaze doesn't run, so you have *something* to reason about beyond a blank cell.
## Licence & attribution
@ -70,7 +73,7 @@ Our own work (every field *except* the failure-rate ones). Licensed under [Creat
### 2. Failure-rate data - Backblaze's terms (NOT CC BY 4.0)
The fields `afr_pct`, `reliability_drive_count`, `reliability_drive_days`, `reliability_failures` (in `drives-2026-06-16.csv` / `drives-2026-06-16.json`) and all of `backblaze-afr-2026-06-16.csv` come from Backblaze Drive Stats.
The fields `afr_pct`, `reliability_drive_count`, `reliability_drive_days`, `reliability_failures` (in `drives-2026-06-17.csv` / `drives-2026-06-17.json`) and all of `backblaze-afr-2026-06-17.csv` come from Backblaze Drive Stats.
- **You can:** use it for free, and build (and even sell) derivative works from it.
- **You must:** credit Backblaze as the source.
@ -80,3 +83,8 @@ The fields `afr_pct`, `reliability_drive_count`, `reliability_drive_days`, `reli
## Updates
This repo is a point-in-time snapshot; the date in each filename is its export date and it can fall behind. For current data use [nasdisks.com/data/](https://www.nasdisks.com/data/) or the endpoints above, which always reflect the latest catalogue and Backblaze figures.
## Changelog
- **2026-06-17** - Added per-drive noise (idle / seek A-weighted sound power, dB) and helium-fill fields. Corrected WD40EZAX / WD60EZAX to CMR (the newer EZAX revision is CMR; only the older EZAZ is SMR).
- **2026-06-16** - Initial public dataset: full specs, CMR/SMR classification, and full-year 2025 Backblaze annualized failure rates.