Dataset · Q4 2025 · computed from raw Backblaze files
Which models make up the 6+ year AFR buckets, Backblaze Q4 2025
Splits the older power-on age buckets (6 to under 7, 7 to under 8, 8 or more years) of Backblaze's Q4 2025 fleet (October 1 to December 31, 2025) by drive model, from the raw daily Drive Stats files. Sample: hard drives of 3 TB or more with a power-on-hours value; 1,424,836, 226,691 and 1,383,760 drive days in the three buckets, with 101, 44 and 70 failures. In the 7 to under 8 year bucket (7.08% AFR), two models hold 27 plus 13 of the 44 failures: HGST HUH728080ALE600 and Toshiba MG07ACA14TA. Association only, one fleet and one quarter.
Download CSV36 rows · 10 columns · CSVPopulation and method
Population: hard drives of 3 TB or more in Backblaze’s own data centers, from the 92 daily files for October 1 to December 31, 2025, using drive days whose power-on hours (SMART 9) are filled in. This file takes the three oldest age buckets of the AFR by age datasetand splits each by drive model. 6 to under 7 years: 1,424,836 drive days, 101 failures, 13 models; 7 to under 8 years: 226,691 drive days, 44 failures, 11 models; 8 or more years: 1,383,760 drive days, 70 failures, 9 models. The build checks that these totals still equal that dataset, and that the model rows add up to each total.
Age buckets use each day’s power-on hours divided by 8,766 hours per year. AFR is failures ÷ drive days × 365 × 100 with an exact two-sided 95% Poisson interval. “Drives” counts distinct serial numbers seen in the bucket during the quarter, so a drive that crosses a bucket boundary is in both.
Chart
Counted here
- 6 to under 7 years: HGST HUH721212ALN604 has 69.3% of the failures (70 of 101) and 54.8% of the drive days. The bucket is 2.59%. Without that model it is 1.76% (31 failures in 643,669 drive days).
- 7 to under 8 years: HGST HUH728080ALE600 has 61.4% of the failures (27 of 44) and 19.7% of the drive days. The bucket is 7.08%. Without that model it is 3.41% (17 failures in 181,989 drive days).
- 8 or more years: ST8000NM0055 has 70.0% of the failures (49 of 70) and 80.7% of the drive days. The bucket is 1.85%. Without that model it is 2.87% (21 failures in 267,224 drive days).
- 7 to under 8 years, the thin bucket: HGST HUH728080ALE600 (22.05%) and TOSHIBA MG07ACA14TA (5.25%) hold 40 of the 44 failures. Without those two the bucket is 1.59% (4 failures in 91,683 drive days).
Limits
- One operator’s fleet, one quarter. Association only: a model’s rate in an age bucket mixes the model’s design with the batches, locations and workloads of the drives that happen to be in that bucket.
- Removing models after seeing which ones failed most is a description of where the failures sit, not a corrected rate for the age bucket.
- Models with few drive days have intervals that run into the tens or hundreds of percent. They are in the file so the shares add up, not because their rates are informative.
- Power-on hours are a drive-reported counter and are used as reported. The intervals assume Poisson failures at a constant rate in each model and bucket and are probably too narrow.
Table
| Age (years) | Model | Drives | Drive days | Days share (%) | Failures | Failures share (%) | AFR (%) | 95% low | 95% high |
|---|---|---|---|---|---|---|---|---|---|
| 6 to under 7 | HGST HUH721212ALN604 | 9,289 | 781,167 | 54.8 | 70 | 69.3 | 3.27 | 2.55 | 4.13 |
| 6 to under 7 | ST12000NM0008 | 6,107 | 289,866 | 20.3 | 14 | 13.9 | 1.76 | 0.96 | 2.96 |
| 6 to under 7 | HGST HUH721212ALE600 | 1,488 | 136,724 | 9.6 | 2 | 2.0 | 0.53 | 0.06 | 1.93 |
| 6 to under 7 | TOSHIBA MG07ACA14TA | 2,955 | 108,079 | 7.6 | 7 | 6.9 | 2.36 | 0.95 | 4.87 |
| 6 to under 7 | ST12000NM0007 | 848 | 77,096 | 5.4 | 7 | 6.9 | 3.31 | 1.33 | 6.83 |
| 6 to under 7 | ST8000NM0055 | 282 | 21,263 | 1.5 | 1 | 1.0 | 1.72 | 0.04 | 9.56 |
| 6 to under 7 | ST8000DM002 | 69 | 5,822 | 0.4 | 0 | 0.0 | 0.00 | 0.00 | 23.13 |
| 6 to under 7 | ST16000NM001G | 37 | 1,837 | 0.1 | 0 | 0.0 | 0.00 | 0.00 | 73.30 |
| 6 to under 7 | HGST HUH721010ALE600 | 20 | 1,276 | 0.1 | 0 | 0.0 | 0.00 | 0.00 | 105.52 |
| 6 to under 7 | ST10000NM0086 | 10 | 761 | 0.1 | 0 | 0.0 | 0.00 | 0.00 | 176.93 |
| 6 to under 7 | HGST HUH728080ALE600 | 5 | 423 | 0.0 | 0 | 0.0 | 0.00 | 0.00 | 318.31 |
| 6 to under 7 | HGST HUH728080ALE604 | 3 | 276 | 0.0 | 0 | 0.0 | 0.00 | 0.00 | 487.84 |
| 6 to under 7 | ST8000DM005 | 11 | 246 | 0.0 | 0 | 0.0 | 0.00 | 0.00 | 547.33 |
| 6 to under 7 | ALL 13 MODELS | 21,124 | 1,424,836 | 100.0 | 101 | 100.0 | 2.59 | 2.11 | 3.14 |
| 7 to under 8 | TOSHIBA MG07ACA14TA | 1,014 | 90,306 | 39.8 | 13 | 29.5 | 5.25 | 2.80 | 8.99 |
| 7 to under 8 | HGST HUH721212ALN604 | 2,134 | 72,180 | 31.8 | 2 | 4.5 | 1.01 | 0.12 | 3.65 |
| 7 to under 8 | HGST HUH728080ALE600 | 931 | 44,702 | 19.7 | 27 | 61.4 | 22.05 | 14.53 | 32.08 |
| 7 to under 8 | ST8000NM0055 | 137 | 8,979 | 4.0 | 0 | 0.0 | 0.00 | 0.00 | 15.00 |
| 7 to under 8 | ST8000DM002 | 94 | 6,238 | 2.8 | 2 | 4.5 | 11.70 | 1.42 | 42.27 |
| 7 to under 8 | ST8000DM005 | 23 | 1,732 | 0.8 | 0 | 0.0 | 0.00 | 0.00 | 77.74 |
| 7 to under 8 | TOSHIBA HDWF180 | 10 | 918 | 0.4 | 0 | 0.0 | 0.00 | 0.00 | 146.67 |
| 7 to under 8 | HGST HUH721010ALE600 | 10 | 563 | 0.2 | 0 | 0.0 | 0.00 | 0.00 | 239.15 |
| 7 to under 8 | HGST HUH728080ALE604 | 6 | 483 | 0.2 | 0 | 0.0 | 0.00 | 0.00 | 278.77 |
| 7 to under 8 | ST12000NM0007 | 4 | 306 | 0.1 | 0 | 0.0 | 0.00 | 0.00 | 440.01 |
| 7 to under 8 | ST10000NM0086 | 4 | 284 | 0.1 | 0 | 0.0 | 0.00 | 0.00 | 474.10 |
| 7 to under 8 | ALL 11 MODELS | 4,367 | 226,691 | 100.0 | 44 | 100.0 | 7.08 | 5.15 | 9.51 |
| 8 or more | ST8000NM0055 | 12,241 | 1,116,536 | 80.7 | 49 | 70.0 | 1.60 | 1.19 | 2.12 |
| 8 or more | ST8000DM002 | 2,117 | 121,218 | 8.8 | 7 | 10.0 | 2.11 | 0.85 | 4.34 |
| 8 or more | ST10000NM0086 | 919 | 83,470 | 6.0 | 12 | 17.1 | 5.25 | 2.71 | 9.17 |
| 8 or more | HGST HUH728080ALE600 | 874 | 38,048 | 2.7 | 0 | 0.0 | 0.00 | 0.00 | 3.54 |
| 8 or more | HGST HMS5C4040BLE640 | 187 | 17,151 | 1.2 | 1 | 1.4 | 2.13 | 0.05 | 11.86 |
| 8 or more | HGST HUH728080ALE604 | 63 | 5,677 | 0.4 | 1 | 1.4 | 6.43 | 0.16 | 35.82 |
| 8 or more | ST12000NM0007 | 8 | 736 | 0.1 | 0 | 0.0 | 0.00 | 0.00 | 182.94 |
| 8 or more | HGST HMS5C4040ALE640 | 8 | 736 | 0.1 | 0 | 0.0 | 0.00 | 0.00 | 182.94 |
| 8 or more | ST8000DM005 | 4 | 188 | 0.0 | 0 | 0.0 | 0.00 | 0.00 | 716.19 |
| 8 or more | ALL 9 MODELS | 16,421 | 1,383,760 | 100.0 | 70 | 100.0 | 1.85 | 1.44 | 2.33 |
Columns
- age_years
- Power-on age bucket in years, from smart_9_raw hours divided by 8766 on each drive day: 6 to under 7, 7 to under 8, or 8 or more.
- model
- Drive model string as it appears in the daily files, or the ALL row that closes each bucket and equals the sum of its model rows.
- drives
- Distinct serial numbers of the model seen in the bucket between 2025-10-01 and 2025-12-31. A drive that crosses a bucket boundary in the quarter counts in both buckets; the ALL row adds the model rows.
- drive_days
- Daily rows of the model that fell in the bucket. This is the exposure.
- share_of_bucket_drive_days_percent
- The model's share of the bucket's drive days, one decimal.
- failures
- Rows with failure = 1 for the model in the bucket.
- share_of_bucket_failures_percent
- The model's share of the bucket's failures, one decimal.
- afr_percent
- failures / drive_days * 365 * 100, two decimals. Computed here.
- afr_low95
- Lower bound of the exact two-sided 95% Poisson (Garwood) interval, in percent.
- afr_high95
- Upper bound of the same interval, in percent. Very wide for models with few drive days.
License
Derivative table computed from the Backblaze Drive Stats daily files for Q4 2025. Cite Backblaze if you use it. You are solely responsible for how you use it. Backblaze's dataset page allows sale of derivative works and forbids selling the Drive Stats data itself. This CSV is a 36-row aggregate, not the daily SMART archive, and that archive is not for sale.
Backblaze Drive Stats use terms · Hard Drive Test Data page