60TB hard drives set to arrive by 2028, industry forecast predicts

zohaibahd

Posts: 934   +19
Staff
Forward-looking: Going from today's 30TB to 60TB in just four years would be a heck of an increase after HDD capacity growth stalled for a while. However, that's exactly what the new International Roadmap for Devices and Systems Mass Data Storage report by IEEE is predicting. The roadmap calls for 60TB hard disk drives to hit the market in 2028.

The increase will be brought about by some advanced new recording innovations, including energy-assisted magnetic recording (EAMR) technologies such as HAMR: Heat Assisted Magnetic Recording, which was pioneered by Seagate, is expected to be a game-changer in supercharging areal densities on spinning disk platters.

Speaking of areal density, that key spec will have to more than double from around 2TB per square inch in 2025 to over 4TB per square inch in 2028 to facilitate 60TB HDDs, the experts say. By 2037, they're projecting over 10TB per square inch, which could yield mind-boggling 100TB hard drives if the other pieces fall into place. However, Seagate previously predicted they'd be able to hit those numbers as early as 2030.

Achieving these densities won't be easy, though. The roadmap states that "critical advances are required" in materials for both the disk media and the read/write heads. This includes the likes of new magnetic and non-magnetic films, sub-1nm thin dielectric coatings for the heads, and improved disk overcoats. The report also sees traditional 7200RPM speeds sticking around so improvements will have to be made in other areas.

Beyond capacities, one peculiar revelation in the roadmap is that global HDD unit sales could surge over the coming years, rather than continuing their decline (except for the slight recent uptick). The forecast predicts sales rising from 166 million in 2022 to 208 million in 2028 and a whopping 359 million annually by 2037.

Beyond HDDs, IEEE also has some interesting insights for the 3D NAND flash market. They note that the path to higher densities/lower costs will involve increasing the bits-per-cell (QLC to PLC), minimizing layer count growth, shrinking lithography, and maintaining high aspect ratio hole etching. But layer scaling will eventually hit a limit, perhaps around 1,000 layers based on current projections.

The report also mentions magnetic tape storage, noting that the market is expected to sustain its recent 30-40% annual capacity growth rate. However, advances in read heads, servo tech, error correction, and smoother media coatings will be required to achieve higher capacities.

Permalink to story:

 
So…. Based on advances that haven’t happened yet - and that may not happen for ages (if ever), we’re expecting 60tb… pardon me if I don’t leap up and cheer…
 
So…. Based on advances that haven’t happened yet - and that may not happen for ages (if ever), we’re expecting 60tb… pardon me if I don’t leap up and cheer…

Can't see why we won't get 60Tb by end of decade at most, I mean 30Tb is pretty impressive. Most of this is just refinement and combining techs. Not a whole new tech. Plus they must have so many new modelling tools and ways to shortened cycle to end product now. Maybe TSMC also benefits from things like better purity of creating base substrate which they use.
most of these things are multiplicative , 5% here 15% there. Seems easy gains with platters is harder now. Add in software error correcting algorithms etc
 
Can't see why we won't get 60Tb by end of decade at most, I mean 30Tb is pretty impressive. Most of this is just refinement and combining techs. Not a whole new tech. Plus they must have so many new modelling tools and ways to shortened cycle to end product now. Maybe TSMC also benefits from things like better purity of creating base substrate which they use.
most of these things are multiplicative , 5% here 15% there. Seems easy gains with platters is harder now. Add in software error correcting algorithms etc
Well, the article pointed out how Seagate thought they’d hit a certain value by 2030 but now it’s been pushed back by years…. So why should we believe that 4 years will bring 60tb? The “leap” from 16 to 22tb took about 5 years… I would like to see some evidence before I declare 60tb a sure thing by 2030…
 
Last edited:
One thing that worries me about these ever increasing data densities, what effect does that have on data life span if you're looking for something for archival storage?
 
Well, the article pointed out how Seagate thought they’d hit a certain value by 2030 but now it’s been pushed back by years…. So why should we believe that 4 years will bring 30tb? The “leap” from 16 to 22tb took about 5 years… I would like to see some evidence before I declare 30tb a sure thing by 2030…
Seagate are already advertising 32TB enterprise drives here: https://www.seagate.com/au/en/products/enterprise-drives/exos-x/x-mozaic/

I've not seen or heard of one in the wild yet, but then I don't hang around data centres either.
 
Even if they don't reach those projections, it's still pretty impressive that there's still innovations in mechanical hard drives. Many people thought they'd be gone long ago after SSD's came out. But here we are years later and HDDs are still going strong. Though I haven't put one in a new build in a while personally.
 
Just waiting on those high capacity prices to come down some and then I'll be happy. I collect a LOT of media (Plex server) and just 2 large capacity mirrored drives would make my day (as apposed to the 6 drives I currently use)...
(just added some awesome oldies - The Hammer House of Horror movies from the 60's and 70's. Peter Cushing was very much a childhood favourite and Sir Christopher Lee started looooong before Star Wars kiddies... LOL)
 
And a steep price for the 60TB hdd.
...and by that time, I'll finally can afford a meager 10TB.
 
Even if they don't reach those projections, it's still pretty impressive that there's still innovations in mechanical hard drives. Many people thought they'd be gone long ago after SSD's came out. But here we are years later and HDDs are still going strong. Though I haven't put one in a new build in a while personally.
Most consumer PCs don't use HDDs. Servers do.
 
True. They still make WD blue and black drives, but I'm not sure who's buying them lol.
I think WD Blue is for surveillance (recording video), Black I don't know whos buying, makes little sense to buy a "high-end" HDD these days. If you need performance you get SSD.
WD Red is good for NAS/Servers.
 
So…. Based on advances that haven’t happened yet - and that may not happen for ages (if ever), we’re expecting 60tb… pardon me if I don’t leap up and cheer…

This is the purpose of these talks - experts who are seeing the cutting edge advising us on what is coming down the pipeline. All of these things are in some way being explored, some are more likely to come, others will take longer or not happen. All of this is OK.
 
This is the purpose of these talks - experts who are seeing the cutting edge advising us on what is coming down the pipeline. All of these things are in some way being explored, some are more likely to come, others will take longer or not happen. All of this is OK.
Except the qualifications to become an "expert" seem hazy at best... and they're often wrong... when it comes to predicting the future, NOONE is perfect... but... I'd prefer if they at least made predictions based on actual evidence instead of conjecture.
 
Can't see why we won't get 60Tb by end of decade at most, I mean 30Tb is pretty impressive. Most of this is just refinement and combining techs. Not a whole new tech. Plus they must have so many new modelling tools and ways to shortened cycle to end product now. Maybe TSMC also benefits from things like better purity of creating base substrate which they use.
most of these things are multiplicative , 5% here 15% there. Seems easy gains with platters is harder now. Add in software error correcting algorithms etc
Seagate has told this multiple time before. Each time with lofty goals and big numbers - only to have none of it materialize.
Current capacity increase tempo is roughly 2TB per year. Current highest capacity HDD normal person can buy is 24TB.

In order to reach 60TB by 2028 the rate of capacity increase needs to be 8TB per year and maintained over four years.
That would be quadruple the rate of today. Extremely unlikely.
As it stands this I far more likely:
2025: 26TB
2026: 28TB
2027: 30TB
2028: 32TB
2029: 34TB
2030: 36TB

FYI with this they will reach 60TB by 2042, not 2028.

Seagate are already advertising 32TB enterprise drives here: https://www.seagate.com/au/en/products/enterprise-drives/exos-x/x-mozaic/

I've not seen or heard of one in the wild yet, but then I don't hang around data centres either.
Unless a regular Joe can buy it then it might as well not exist. The biggest a regular consumer can buy is still 24TB.
Maybe make 26TB, 28TB and 30TB available before advertising unobtanium 32TB Seagate!
 
Unless a regular Joe can buy it then it might as well not exist. The biggest a regular consumer can buy is still 24TB.
Maybe make 26TB, 28TB and 30TB available before advertising unobtanium 32TB Seagate!
Technology existing and being available to a regular consumer are not mutually inclusive.

I don't have access or funding to buy a rocket capable of reaching space, doesn't mean they don't exist. :p
 
Technology existing and being available to a regular consumer are not mutually inclusive.

I don't have access or funding to buy a rocket capable of reaching space, doesn't mean they don't exist. :p
Except this 32TB model is not even available for enterprise at the moment.
Lets leave rockets out of this. We're still talking about technology that is within price range for most people. Meaning less that 700€ most likely for this 32TB model. Current 24TB models are around 500€.
 
Seagate has told this multiple time before. Each time with lofty goals and big numbers - only to have none of it materialize.
Current capacity increase tempo is roughly 2TB per year. Current highest capacity HDD normal person can buy is 24TB.

In order to reach 60TB by 2028 the rate of capacity increase needs to be 8TB per year and maintained over four years.
That would be quadruple the rate of today. Extremely unlikely.
As it stands this I far more likely:
2025: 26TB
2026: 28TB
2027: 30TB
2028: 32TB
2029: 34TB
2030: 36TB

FYI with this they will reach 60TB by 2042, not 2028.


Unless a regular Joe can buy it then it might as well not exist. The biggest a regular consumer can buy is still 24TB.
Maybe make 26TB, 28TB and 30TB available before advertising unobtanium 32TB Seagate!
Yeah went back and checked article . Quite telling line : Critical advances needed. Ie 3 or 4 mentioned, suggested even more. So yeah a bit optimistic of them. Key point, to get to 60Gb need to double density, so could get a big jump if they crack that
 
Back