Seagate's Roadmap Shows How We Can Expect 100 TB HDDs in 2025

[dmartin]

Do you know what's better than a 10 TB HDD? A 100 TB HDD. Seagate has published a product roadmap showing the evolution of their HDDs in the next few years. According to that roadmap, we'll be able to buy a 100 TB hard drive in 2025. Other makers such as Western Digital talked about 40 TB drives in 2025 thanks to MAMR (Microwave Assisted Magnetic Recording) technology, but it seems Seagate is even more optimistic, and in fact they'll use HDMR (Heated Dot Magnetic Recording), an evolution of the HAMR (Heat Assisted Magnetic Recording) technology.

Seagate expects to have 20 TB HAMR-HDDs by 2020, and from that moment on they'll continue to work on this technology to launch 36 TB HDDs by 2022, 48 TB drives before 2024 and that 100 TB units in 2025. HAMR will enable double areal density growth each 2.5 years according to Seagate. The current PMR (Perpendicular Magnetic Recording) technology could disappear in that future, and it seems HDDs will continue to be a valuable asset for those who look for the cheapest price per GB.



View at TechPowerUp Main Site

 

[dj-electric]

What does better for the environment? A 100TB HDD in 2025 or a 200TB SSD by then?

Seriously asking. It seems that the HDD cartel is forcefully refusing to let go and help the NAND market grow as intended.
These shouldn't exist. HDDs in 2025 shouldn't exist.

 

[RCoon]

3.5" is not a good enough form factor for HDD growth anymore.

 

[Dante Uchiha]

What does better for the environment? A 100TB HDD in 2025 or a 200TB SSD by then?

Seriously asking. It seems that the HDD cartel is forcefully refusing to let go and help the NAND market grow as intended.
These shouldn't exist. HDDs in 2025 shouldn't exist.

I imagine it is difficult to happen if the only way to reduce the SSDs cost is by reducing the lifespan.

It is strange to call "HDD cartel", in a market where the price of Ram and Nand is artificially manipulated.

 

[kastriot]

It's time for dead horse to rest in 5 years and not reviving it like some horsenstein.

 

[NC37]

I imagine it is difficult to happen if the only way to reduce the SSDs cost is by reducing the lifespan.

It is strange to call "HDD cartel", in a market where the price of Ram and Nand is artificially manipulated.

And RAM producers just got class actioned again for price fixing after they got caught doing it back in the late 90s/early 2000s. Still remember the day I got my settlement check and I had to take awhile to remember what it was about.

 

[Frick]

What does better for the environment? A 100TB HDD in 2025 or a 200TB SSD by then?

Seriously asking. It seems that the HDD cartel is forcefully refusing to let go and help the NAND market grow as intended.
These shouldn't exist. HDDs in 2025 shouldn't exist.

Why wouldn't they? They are still good for many, many things. They're simpler to manufacture and afaik doesn't use as much exotic materials so I see no reason why they can't exist side by side. I mean for some applications optical storage is still prefered.

 

[DeathtoGnomes]

$1 per terabyte drives, heh.

How long has it been since it was $1 per megabyte?

 

[holyprof]

What does better for the environment? A 100TB HDD in 2025 or a 200TB SSD by then?

Seriously asking. It seems that the HDD cartel is forcefully refusing to let go and help the NAND market grow as intended.
These shouldn't exist. HDDs in 2025 shouldn't exist.

Although I agree with you that in 2025 no end-user computer will have a mechanical hard drive, it's not the same in the enterprise market.

About the cartel, are you telling us that "WD and Seagate" cartel are stopping the NAND cartel (Samsung, Micron, whoever bought Toshiba(WD) and SK Hynix, probably missed 1 or 2) from making cheap and reliable 100TB drives?
If NAND storage is so superior to mechanical drives, it will naturally eliminate it (and it will, but not before 2025). As much as I would like to recycle my 4TB WD greens and replace them with quieter, cooler and faster SSDs, NAND memory is hitting a wall and you can't expect prices to drop as they have been doing in the last 10 years.

*Written on my main pc with no mechanical HDD in it *

 

[arroyo]

Seagate... nope.
I wouldnt buy this crap for my data even if it would cost 1$ per 1TB.
I am working in data recovery for last 5 years and I see 8 out of 10 HDD are physically damaged Seagates. They cannot even design good head parking mechanism for heads. Their disks comparing to Toshiba or Hitachi are design to fail.

 

[R0H1T]

I imagine it is difficult to happen if the only way to reduce the SSDs cost is by reducing the lifespan.

It is strange to call "HDD cartel", in a market where the price of Ram and Nand is artificially manipulated.

You could say the same about (some) CPUs & GPUs but the thing with HDD is that they've kinda hit rock bottom, so the prices shouldn't be as high as they are. I predict mechanical storage to be extinct by the end of the next decade if alternatives like Optane, memristor grow the way NAND has done. They'll be the tape storage of our generation, RIP tapes & Vinyl

 

[rtwjunkie]

Seriously asking. It seems that the HDD cartel is forcefully refusing to let go and help the NAND market grow as intended.

You don’t realize that WD has their own SSD production, do you, with this HDD cartel nonsense?

Although I agree with you that in 2025 no end-user computer will have a mechanical hard drive, it's not the same in the enterprise market.

Really? No end user? It’s still much cheaper for me to have a 4 TB game drive than a 4TB SSD for the same purpose. It’s still cheaper for my 20 TB (+) media and storage server to use HDD.

Unless a miracle breakthrough comes about, and considering all my server HDD last about 8 years, I expect I will be running with them long past 2025.

So, TL;DR I don’t know how you can make a blanket statement that no end user computer will have an HDD in 2025.

 

[Dante Uchiha]

You could say the same about (some) CPUs & GPUs but the thing with HDD is that they've kinda hit rock bottom, so the prices shouldn't be as high as they are. I predict mechanical storage to be extinct by the end of the next decade if alternatives like Optane, memristor grow the way NAND has done. They'll be the tape storage of our generation, RIP tapes & Vinyl

If they solve the "SSD weaknesses", then there's no problem with that. It would be great to have a reliable and affordable fast storage.

 

[neatfeatguy]

I imagine it is difficult to happen if the only way to reduce the SSDs cost is by reducing the lifespan.

It is strange to call "HDD cartel", in a market where the price of Ram and Nand is artificially manipulated.

Off-topic, but.......Just remember, it's not computer/hardware that's involved in price fixing: https://www.cnbc.com/2018/10/19/starkist-admits-fixing-tuna-prices-faces-100-million-fine.html
Greedy people all around us.

 

[phill]

I just wonder how many more datacenters we'd need to hold these size drives for all the data (we shall call it that) that will be around at that point... Wow.... I wonder if we'd have to have new SATA standards and formatting for that??

 

[TheGuruStud]

Pffft. Drives any larger than today will spend most of their time verifying data. Bits will be flipped.

 

[Fx]

Although I agree with you that in 2025 no end-user computer will have a mechanical hard drive, it's not the same in the enterprise market.

I wouldn't be so sure. I certainly might. It all comes down to cost, capacity, endurance and use case.

Seagate... nope.
I wouldnt buy this crap for my data even if it would cost 1$ per 1TB.
I am working in data recovery for last 5 years and I see 8 out of 10 HDD are physically damaged Seagates. They cannot even design good head parking mechanism for heads. Their disks comparing to Toshiba or Hitachi are design to fail.

Yep! I wholly wrote Seagate off around 2006. I've seen first-hand too many failed drives from them. They would be the last manufacturer I would buy from.

 

[Tomorrow]

Seagate is lying as usual. Forget 100TB HDD by 2025. At best we will have 40TB by then. Seagate themselves have previously said HAMR limit is 40TB. Going from 16TB now to 40TB in ~6 years seems doable if not overly optimistic considering HDD size increases in previous 6 years. 100TB is a clickbait lie - pure and simple.

Personally i'm much more intrested in Western Digital's MAMR alternative. It supposedly resolves HAMR related reliability issues by using microwaves instead of pure heat. Also slated for next year.

 

[Prima.Vera]

This really doesn't matter, it's even hilarious. Even NOW, TODAY, you can easily buy 100TB SSDs, for premium price, true, but nevertheless they exists. Imagine how much their price will drop in 7 years, common.
Just let the rust junk die already. Even at 100TB they will never ever be able to match the speed and reliability of nowadays SSDs.