Dataset · Q1 and Q2 2026 · vendor claim

Backblaze drive model AFR, Q1 versus Q2 2026

Side-by-side published annualized failure rates for 33 hard-drive models in Backblaze's Q1 2026 (January 1 to March 31) and Q2 2026 (April 1 to June 30) Drive Stats model tables. 31 models appear in both. Fleet AFR in the published totals rows rose from 1.24% (1,030 failures, 30,203,180 drive days) to 1.73% (1,498 failures, 31,553,350 drive days). Vendor claim; Backblaze's own fleet, not an industry sample.

Download CSV33 rows · 11 columns · CSV

File /data/datasets/backblaze-q1-q2-2026-afr-shift.csv · JSON metadata · Chart (SVG) · Q2 2026 model table · All datasets · AFR glossary entry

Population and method

Population: hard drives in Backblaze’s own data centers that met the report inclusion rule (drive count above 100 and drive days above 10,000 as of the quarter end). Q1 covers January 1 through March 31, 2026 and lists 33 models. Q2 covers April 1 through June 30, 2026 and lists 31 models. The CSV has 33 rows, one per model that appears in either table. 31 models are in both, and 2 are only in Q1. The two Q1-only models are the HGST HMS5C4040BLE640 and HUH728080ALE600, both with zero Q1 failures. Cells are copied from the printed tables. This page recomputes every AFR as failures ÷ drive days × 365 × 100 and each matches within 0.011 percentage point.

Sample size: the Q1 rows sum to 30,203,180 drive days and 1,030 failures (1.2447%), matching the report’s totals row of 1.24%. The Q2 rows sum to 31,553,350 drive days and 1,498 failures (1.7328%), matching 1.73%. Q2 drive counts sum to the published 354,415.

Chart

Scatter of 31 drive models: published Q1 2026 AFR on the x axis, published Q2 2026 AFR on the y axis, with a dashed line where the two are equal. 23 models are above the line, 8 below. The table below lists every plotted value. — Every point is a row of the model table below. 23 of 31models have a higher published AFR in Q2 than in Q1, 8 lower, and 0 equal. Marker area follows Q2 drive days with a minimum size.

Counted here

From this CSV, not from report prose. Over the 31 models in both tables, pooled AFR was 1.25% in Q1 (1,030 failures, 30,100,401 drive days) and 1.73%in Q2 (1,498 failures, 31,553,350 drive days). Failures rose by 468 in total. The five largest per-model increases in failure count account for 287 of them:

HGST HUH721212ALN604 (12 TB): 95 to 180 failures (+85), AFR 3.98% to 7.63%.
Toshiba MG07ACA14TA (14 TB): 94 to 179 failures (+85), AFR 1.03% to 1.94%.
Seagate ST8000NM0055 (8 TB): 39 to 111 failures (+72), AFR 1.21% to 3.44%.
HGST HUH721212ALE604 (12 TB): 86 to 109 failures (+23), AFR 2.65% to 3.34%.
Toshiba MG10ACA20TE (20 TB): 42 to 64 failures (+22), AFR 0.92% to 1.1%.

Limits

One operator’s fleet, not an industry sample. Rates are vendor claims.
Association only. This file shows that rates differ between two quarters. It does not show why. The published tables carry no confidence interval.
Small counts dominate the noise. 8 of 31 models had fewer than 10 Q2 failures, so a one-failure change moves their AFR by several points.
Q1 drive counts in the rows sum to 346,596. The Q1 report text says 341,263 hard drives remained to analyze, a gap of 5,333. Q1 drive days, failures, and AFR match the totals row, and the Q1 totals row prints no drive count. This file does not decide which Q1 count is right.
Drive count is a snapshot at quarter end. AFR uses drive days, so a model that grew during the quarter (for example the Toshiba MG10ACA20TE or the WDC WUH722626ALE6L4) has a count that is not comparable across quarters one for one.
Vendor claim, Q1 2026 press release: drives that fail before completing their first day in production are not captured in the dataset.
Q1 and Q2 tables are reported on the same inclusion rule but are separate reports. Q2’s own report text mislabels one paragraph as “Q1 2026”; the table title and dates are Q2.

Model table

Vendor claim. Same 33 rows as the CSV. Empty cells mean the model is not in that quarter’s table.
MFG	Model	Size (TB)	Q1 drives	Q1 drive days	Q1 failures	Q1 AFR (%)	Q2 drives	Q2 drive days	Q2 failures	Q2 AFR (%)
HGST	HMS5C4040BLE640	4	186	15,858	0	0
HGST	HUH728080ALE600	8	1,026	86,921	0	0
HGST	HUH721212ALE600	12	2,608	234,350	8	1.25	2,599	236,107	7	1.08
HGST	HUH721212ALE604	12	13,277	1,183,774	86	2.65	13,196	1,190,182	109	3.34
HGST	HUH721212ALN604	12	9,780	870,252	95	3.98	9,694	861,297	180	7.63
Seagate	ST8000DM002	8	8,910	639,473	25	1.43	6,903	563,210	38	2.46
Seagate	ST8000NM000A	8	247	21,552	1	1.69	239	21,499	0	0
Seagate	ST8000NM0055	8	13,232	1,176,619	39	1.21	13,112	1,177,541	111	3.44
Seagate	ST10000NM0086	10	981	86,667	11	4.63	965	86,092	22	9.33
Seagate	ST12000NM0007	12	983	87,389	11	4.59	972	87,496	10	4.17
Seagate	ST12000NM0008	12	18,650	1,664,558	129	2.83	18,521	1,671,057	145	3.17
Seagate	ST12000NM000J	12	1,087	96,887	1	0.38	1,079	97,746	0	0
Seagate	ST12000NM001G	12	13,216	1,185,545	33	1.02	13,177	1,193,856	44	1.35
Seagate	ST14000NM000J	14	433	37,258	1	0.98	497	43,349	0	0
Seagate	ST14000NM001G	14	10,519	944,400	21	0.81	10,498	949,396	39	1.5
Seagate	ST14000NM0138	14	1,251	111,709	15	4.9	1,235	110,410	25	8.26
Seagate	ST16000NM000J	16	129	10,117	1	3.61	171	14,221	1	2.57
Seagate	ST16000NM001G	16	34,729	3,098,105	44	0.52	35,100	3,162,117	56	0.65
Seagate	ST16000NM002J	16	466	41,931	0	0	467	42,086	4	3.47
Seagate	ST24000NM002H	24	9,605	820,090	74	3.29	9,631	861,587	78	3.3
Toshiba	MG07ACA14TA	14	37,263	3,343,724	94	1.03	37,171	3,368,924	179	1.94
Toshiba	MG07ACA14TEY	14	996	88,602	5	2.06	989	89,164	6	2.46
Toshiba	MG08ACA16TA	16	40,036	3,548,250	102	1.05	39,932	3,614,263	100	1.01
Toshiba	MG08ACA16TE	16	6,264	553,878	21	1.38	6,306	568,379	16	1.03
Toshiba	MG08ACA16TEY	16	4,885	435,063	26	2.18	4,857	439,243	44	3.66
Toshiba	MG09ACA16TE	16	506	36,988	1	0.99	905	68,429	2	1.07
Toshiba	MG10ACA20TE	20	20,451	1,665,263	42	0.92	25,270	2,117,524	64	1.1
Toshiba	MG11ACA24TE	24	7,203	495,387	3	0.22	9,620	824,903	13	0.58
WDC	WUH721414ALE6L4	14	8,640	776,557	7	0.33	8,635	783,318	23	1.07
WDC	WUH721816ALE6L0	16	2,994	267,868	24	3.27	2,970	268,875	30	4.07
WDC	WUH721816ALE6L4	16	26,801	2,402,732	64	0.97	26,827	2,432,374	77	1.16
WDC	WUH722222ALE6L4	22	45,638	3,992,942	42	0.38	45,665	4,146,339	63	0.55
WDC	WUH722626ALE6L4	26	3,604	182,471	4	0.8	7,212	462,366	12	0.95

Columns

mfg (string): Manufacturer label as published by Backblaze: HGST, Seagate, Toshiba, or WDC.
model (string): Drive model identifier as published.
size_tb (integer): Nominal capacity in terabytes as published.
q1_drive_count (integer): Drives of this model in the Q1 2026 table. Empty if the model is not in that table.
q1_drive_days (integer): Drive days observed in Q1 2026 (January 1 to March 31, 2026). Empty if not in the table.
q1_failures (integer): Failures in Q1 2026 as published. Empty if not in the table.
q1_afr_percent (number): Q1 2026 AFR in percent, copied as published, not refit. Empty if not in the table.
q2_drive_count (integer): Drives of this model in the Q2 2026 table. Empty if the model is not in that table.
q2_drive_days (integer): Drive days observed in Q2 2026 (April 1 to June 30, 2026). Empty if not in the table.
q2_failures (integer): Failures in Q2 2026 as published. Empty if not in the table.
q2_afr_percent (number): Q2 2026 AFR in percent, copied as published, not refit. Empty if not in the table.

License

Derivative table transcribed from Backblaze Drive Stats reports for Q1 2026 and Q2 2026. 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 small derived comparison, not Backblaze's daily SMART archive, and that archive is not for sale.

Backblaze Drive Stats use terms · Q1 2026 report · Q2 2026 report

Sources

Backblaze Drive Stats for Q1 2026 (accessed 2026-10-09)
Backblaze Drive Stats for Q2 2026 (accessed 2026-10-09)
Hard Drive Reliability & Test Data (accessed 2026-10-09)