In laboratories and hyperscale data halls, the terabyte ceiling on nearline hard drives is being rewritten faster than most UK procurement cycles can track. Seagate confirmed in March 2026 that its Mozaic 4+ HAMR-based drives — a 44TB-class platform built on more than 4TB per disk — are already shipping in volume to two leading hyperscale cloud providers, while Western Digital's rival 40TB UltraSMR ePMR drive remains in customer qualification, with volume production only slated for the second half of 2026. For UK teams weighing petabyte storage server design against near-term budgets, the gap between 'shipping to hyperscalers' and 'orderable through UK distribution' is this year's most consequential distinction.
View the data behind this chart
| Western Digital | Seagate | Toshiba | |
|---|---|---|---|
| Near-term milestone… | TB40 | TB44 | TB40 |
| Longer-term roadmap… | TB100 | TB50 | TB55 |
The State of Play in Mid-2026: Hyperscale Shipping vs UK Channel Reality
As of July 2026, headline 'largest hard drive' claims are being made from two very different vantage points: what hyperscale cloud providers are already receiving in volume, and what vendors say will eventually be broadly orderable. Seagate's confirmation in March 2026 that Mozaic 4+ HAMR-based drives — a 44TB-class platform built on over 4TB per disk — are shipping in volume to two leading hyperscale cloud providers is the highest confirmed shipping status in this dataset. That is a live production fact, not a forecast.
Contrast that with Western Digital's rival 40TB UltraSMR ePMR drive, which the company has said it expects to be the world's highest-capacity HDD at launch. As of February 2026, that drive was still in customer qualification, with volume production planned for the second half of 2026 — the very window this article sits in. Neither of these capacities carries a confirmed UK retail or SME channel listing in the sources reviewed here; both remain hyperscale-qualification products at the time of writing.
For UK archive and backup planners, that distinction matters more than the terabyte number itself. Use current, shipping generations to size real projects via the storage solution finder, and treat 40TB-plus roadmap SKUs as planning inputs until a distributor confirms stock.

Three Vendors, Three Roadmaps: Comparing the Capacity Targets
Western Digital's roadmap runs on two parallel tracks. Its 40TB UltraSMR ePMR drive is heading to volume production in H2 2026, while a separate HAMR line is in qualification with two hyperscale customers, with ramp production planned for 2027 and a scaling path to 100TB by 2029. WD has also said it intends to extend the ePMR line itself to 60TB by borrowing HAMR-derived innovations, without increasing power consumption — a notable claim for operators running dense, always-on nearline arrays.
Seagate is furthest ahead on confirmed shipping status: the 44TB-class Mozaic 4+ platform, built on over 4TB per disk, is already moving in volume to two hyperscale cloud providers. Seagate's own longer-range roadmap projects 50+TB drives and more than 10TB per disk by 2032.
Toshiba's public roadmap is the most conservative of the three. A 40TB-class drive using MAMR with 11 or 12 platters is targeted for 2027 — a year behind WD's stated 2026 volume-production timeline for its own 40TB drive — followed by 45TB in 2028 and 55TB or more after 2029.
Decoding the Technology: HAMR, ePMR/UltraSMR and MAMR
HAMR (heat-assisted magnetic recording) is the technology behind both Seagate's shipping Mozaic 4+ platform and WD's separate, still-qualifying HAMR line. It is the route both vendors are using to push per-disk density higher, which is how Seagate reaches its current 44TB-class capacity and how WD expects to eventually reach 100TB by 2029.
ePMR combined with WD's UltraSMR shingled-writing scheme is the basis of WD's nearer-term 40TB drive, currently in qualification for H2 2026. WD has said this same ePMR approach can be pushed to 60TB by borrowing HAMR-derived improvements — without a stated increase in power draw — which is relevant to anyone stacking dense JBODs for HDD vs QLC vs TLC storage tiering decisions.
Toshiba's chosen near-term path is MAMR, applied to an 11- or 12-platter design to reach its own 40TB class in 2027. A higher platter count is one way to add capacity without yet requiring a full HAMR transition, which likely explains Toshiba's more cautious public timeline relative to WD and Seagate.
Reading the Bandwidth Signal Behind Rising Density
Capacity is not the only number moving. A 2026 roadmap comparison indicated WD's sequential bandwidth gains were projected to start at up to 2x, with a longer-term roadmap extending to 4x, 6x and 8x. In practice, each generational density jump tends to lift sequential throughput alongside capacity, which is the figure that actually determines how long a backup window takes once you've filled a 40TB-class drive rather than a smaller one.
Qualification Status as the Real Availability Signal
The single most useful filter for a 2026 roadmap is qualification status, not the headline terabyte figure. 'Shipping in volume to two hyperscale cloud providers' (Seagate's current position) is a different procurement reality from 'in customer qualification with volume production planned' (WD's position on both its 40TB ePMR drive and its HAMR line). A separate, broader industry projection from January 2026 suggested 40+TB HDD availability was expected during 2026, with 100TB HDDs potentially arriving by the early 2030s — a directionally consistent but independently sourced view that sits alongside, not on top of, the vendor-specific figures above.
£/TB and Power Draw: The TCO Metrics Hyperscale Headlines Don't Show
None of the vendor announcements reviewed here include a published UK £/TB price, and that absence is itself informative. Western Digital's 40TB UltraSMR ePMR drive and Seagate's 44TB-class Mozaic 4+ platform are both, at the time of writing, hyperscale-qualification or hyperscale-shipping products rather than broadly distributed SKUs — and £/TB, the metric most UK organisations actually use to compare storage options, only becomes meaningful once a drive is orderable through normal channels rather than qualified for or shipping to two named cloud customers. A 40TB-class drive can only materially change UK archive economics once it becomes broadly orderable, not merely announced, so buyers should expect the effective £/TB on these drives to improve as they move from hyperscale-only qualification toward wider channel availability, even though none of the vendor statements gives a firm date for when that happens.
The one specific power claim in this roadmap comes from Western Digital, which says it can extend its ePMR line from 40TB to 60TB by borrowing HAMR-derived innovations without increasing power consumption — meaning a UK buyer could, on paper, get 50% more capacity per drive at the same power draw, directly improving power-per-TB and therefore total cost of ownership for archive and backup racks. Seagate and Toshiba's public roadmaps in this dataset do not include an equivalent power-consumption claim for their HAMR and MAMR platforms respectively, so UK data centre and archive planners should treat power-per-TB as an open question for those two vendors' upcoming generations, alongside rack density and supply certainty, rather than assuming parity with WD's stated position.
View the data behind this chart
| Capacity (TB) | 2026 | 2028 | 2029 |
|---|---|---|---|
| WD capacity roadmap | 40 | 60 | 100 |
What This Means for UK Archive and Backup Buyers
UK channel availability tends to lag vendor qualification milestones even once headline capacities are public. A 40TB-class drive only changes archive economics once it is broadly orderable — not merely announced — so current buying decisions are better made by separating nearline HDDs for active backup repositories from true long-retention archive tiers, an approach explored in more depth in our all-flash vs nearline HDD TCO analysis.
For public-sector and regulated workloads in particular, compliance factors — immutable backup, data residency, procurement through established distributors — will typically matter more than theoretical top-end capacity. Treat HAMR adoption specifically as a roadmap risk to be monitored, not a variable to plan around, until product SKUs are confirmed shipping through UK channels.
The Long View: Roadmap Beyond 2026
Looking past this year's qualification milestones, WD's stated scaling path runs from 40TB in 2026 to 60TB in 2028 and 100TB in 2029, with the 40TB-to-60TB leg using ePMR extended with HAMR-derived techniques and the 100TB milestone tied to WD's separate HAMR line, which is due to begin ramp production in 2027. Seagate's roadmap extends to 50+TB drives and over 10TB per disk by 2032, with some projections indicating 100TB-class products in the early 2030s. Toshiba's public roadmap sits between the two nearer-term dates, moving from 40TB (2027) to 45TB (2028) to 55TB-plus after 2029.
None of these later-stage figures should be read as confirmed consumer or SME availability dates — they are vendor-stated engineering and qualification targets, several years apart in some cases, and worth revisiting as each milestone is reached rather than treated as fixed.
Bottom Line for UK Buyers in Mid-2026
The confirmed hyperscale ceiling today is Seagate's 44TB-class Mozaic 4+ platform, shipping in volume but only to two named cloud providers. The nearest broadly relevant milestone is WD's 40TB UltraSMR ePMR drive, with volume production planned for this half of 2026. True 100TB-class drives remain a multi-year roadmap item tied to HAMR ramp production starting in 2027. Buy current, shipping generations for live projects now, and track this roadmap for the point at which any of these named capacities actually reaches UK distribution.
Sources
Every figure in this article traces to the sources below.
- •Western Digital — 40TB UltraSMR ePMR, HAMR qualification, 60TB extension and 100TB roadmap
- •Forbes (March 2026) — Seagate Mozaic 4+ HAMR drives shipping in volume to hyperscale customers
- •TechRadar (Feb 2026) — Toshiba's 40TB MAMR and longer-term capacity roadmap
- •Blocks & Files (Feb 2026) — WD 40TB drive qualification and bandwidth roadmap detail
- •TechRadar (Dec 2025) — WD's stated 40TB-to-100TB capacity scaling path by year
- •Forbes (Dec 2025) — Industry-wide projection on 40+TB and 100TB HDD timing
View the data behind this chart
| Status Mid-2026 | Qualification… | Roadmap Target | |
|---|---|---|---|
| Seagate Mozaic 4+ HAMR | Shipping in volume | 2 hyperscale customers | 50+TB by 2032 |
| WD UltraSMR ePMR | In qualification | H2 2026 volume | Extends to 60TB |
| WD HAMR | In qualification | 2 hyperscale customers | 100TB by 2029 |
| Toshiba MAMR | Roadmap stage | Targeted for 2027 | 55TB+ after 2029 |
The 9 verified data points behind this study are free to download and reuse with attribution (CC BY 4.0).
Cite as: Servnet Research, “Largest Hard Drive 2026: HDD Roadmap Tracker (UK)”, servnetuk.com, 2026.