Wednesday, October 28th 2020

Seagate: 20 TB HAMR Drives Arrive in December, 50 TB Capacities in 2026

In its latest earnings call, Seagate, a manufacturer of high-capacity drives, has revealed several interesting points about its upcoming releases of next-generation hard drives. More specifically, the company has disclosed a shift to a new generation of HDDs based on so-called heat-assisted magnetic recording (HAMR) technology. This technology is set to bring many improvements compared to the one currently used by Seagate's rivals like Western Digital. The rivaling company uses energy-assisted perpendicular magnetic recording (ePMR) and microwave-assisted (MAMR) technologies and it already has a 20 TB drive in the offering. Seagate announced that they will unveil a 20 TB HDD in December this year, with the use of HAMR technology, which will bring many improvements like better speed and more efficient disk read/write.

"We remain on track to ship 20-TB HAMR drives starting in December, which is an important milestone, as we believe HAMR technology will be the industry's path to scaling a real density and increasing drive capacities," said Dave Mosley, CEO of Seagate. "Seagate will be the first to ship this crucial technology with a path to deliver 50-TB HAMR drives forecast in 2026."
Source: Seagate Earnings Call (Transcript)
Add your own comment

60 Comments on Seagate: 20 TB HAMR Drives Arrive in December, 50 TB Capacities in 2026

#26
windwhirl
owen10578Unless they can read in GB/s speeds they will take forever to fill a 50TB drive, I don't see why would any typical home user would need one.
It seems unlikely that these drives will be used by home users. Frankly, seems more directed at enterprise markets.
Posted on Reply
#27
Mindweaver
Moderato®™
Here I am just wanting an affordable 10-12tb drive. When I say affordable I mean less than 100 bucks.
Posted on Reply
#28
Rei
owen10578Unless they can read in GB/s speeds they will take forever to fill a 50TB drive, I don't see why would any typical home user would need one.
At the time of release, it will be more for the prosumer market. It will take prolly some years after until it can be affordable enough for average consumer.
Also, the typical use case would to not fill it right away from the get-go but instead, it would be filled throughout the course of it's storage needs which is usually in years.
windwhirlIt seems unlikely that these drives will be used by home users. Frankly, seems more directed at enterprise markets.
That & the prosumer market. Eventually the consumers might be able to afford one a decade down the line.
Posted on Reply
#29
Hattu
Caring1It's HAMR time.
Thanks a lot! Most annoying earworm for a long time! Now i must wash my ears with some proper black metal! 1-3 hours should do the Job.

:roll::roll::roll:
Posted on Reply
#30
windwhirl
MindweaverHere I am just wanting an affordable 10-12tb drive. When I say affordable I mean less than 100 bucks.
Those start at +170 at least... Wait a few more years. HDD GB cost has been dwindling but no longer as fast as years past...
ReiThat & the prosumer market. Eventually the consumers might be able to afford one a decade down the line
Will they want too, though? I mean, I see myself doing that, perhaps (I'm a digital hoarder, so it is to be expected :laugh: ), but I don't think most people need more than 4 TB today...

And then there is the SSD GB cost to consider. It's been going down quite a bit in these last years, and I kinda expect that trend to continue for years to come, so it seems to me that in a decade most home users will simply have one big SSD for most things, while throwing everything else to the Internet (be that social media networks or cloud storage like OneDrive or Google Drive). Well, if the whole game streaming service thing doesn't take off at all, make that two big SSDs :D
Posted on Reply
#31
dragontamer5788
chris.londonImagine rebuilding a 6 disk RAID-Z2/RAID 6 array...
At these capacities, you probably would just RAID1 or RAID10 these drives together.
Posted on Reply
#32
Mindweaver
Moderato®™
Reithe typical use case would to not fill it right away from the get-go but instead, it would be filled throughout the course of it's storage needs which is usually in years.
This could be enough headroom to future proof the pictures my wife takes... lol Yeah this is just over time storage. If you tried filling it up completely with one straight dump it would be like trying to fill an 80gb drive over usb1... lol
Posted on Reply
#33
Rei
windwhirlWill they want too, though? I mean, I see myself doing that, perhaps (I'm a digital hoarder, so it is to be expected :laugh: ), but I don't think most people need more than 4 TB today...

And then there is the SSD GB cost to consider. It's been going down quite a bit in these last years, and I kinda expect that trend to continue for years to come, so it seems to me that in a decade most home users will simply have one big SSD for most things, while throwing everything else to the Internet (be that social media networks or cloud storage like OneDrive or Google Drive). Well, if the whole game streaming service thing doesn't take off at all, make that two big SSDs :D
LOL, like you, I too, am a digital hoarder, though I haven't accumulated as much. Maybe if I had tidy up my messy storages, I would have freed up even more space. Anyway, I can see a decade later that digital files would have inflated in size to meet the demands in hardware advancement, which would make a 10TB drive pretty tempting.

Sure SSD pricing have dropped considerably to being easily affordable but it seems that HDD is also dropping in price while still being considerably larger in capacity & longevity.
Mindweaverlol Yeah this is just over time storage. If you tried filling it up completely with one straight dump it would be like trying to fill an 80gb drive over usb1... lol
LOL, I just did that earlier this month except it was 60GB HDD PATA-to-USB over a USB 1.1 on a 2002 laptop (which is why there is no USB 2.0). To be fair though, this was planned to be done overnight which must've took nearly 20 hours, I think.
Posted on Reply
#34
Metroid
kayjay010101SSD prices are plummeting much faster than HDDs.
Much faster, by 2026 I see hdd close to death unless it develops something incredible to keep the gap between ssd's and hdd's because the way things are, ssd's will become cheaper than hdd's, enterprises are already turning to ssd's because is a lot faster and responsive, reason ssd's prices are plummeting, I mean buying a 64tb for 10k is incredible for enterprises, to get the same storage and slower on hdd's, they need to buy 4 x 18tb $500 each = 2k, plus hdd's are not reliable like ssd's, hdd's die a lot compared to ssd's.
Posted on Reply
#35
okbuddy
26tb max

edit: they said there will be 100tb drive in 2020
Posted on Reply
#36
Tomorrow
okbuddy26tb max

edit: they said there will be 100tb drive in 2020
Indeed they did. And we are 1/5th or 20% of that. So they overestimated by 80%. I believe that was also Seagate. I dont see WD making such outlandish claims.
Posted on Reply
#37
windwhirl
MetroidMuch faster, by 2026 I see hdd close to death unless it develops something incredible to keep the gap between ssd's and hdd's because the way things are, ssd's will become cheaper than hdd's, enterprises are already turning to ssd's because is a lot faster and responsive, reason ssd's prices are plummeting, I mean buying a 64tb for 10k is incredible for enterprises, to get the same storage and slower on hdd's, they need to buy 4 x 18tb $500 each = 2k, plus hdd's are not reliable like ssd's, hdd's die a lot compared to ssd's.
I definitely believe enterprises that need to store a few hundreds of TBs at most will probably move completely to SSDs by next decade. Giant storage services (Dropbox, Google Drive and others) will probably migrate a sizable amount of their storage to SSDs, if they haven't already, although they will keep HDDs around for a while.
Posted on Reply
#38
Ashtr1x
Reliability matters, I would like to see these come to market quick. WD with their superb HGST technology / Ultrastar vs Seagate whoever brings these first they win. As for SSD vs HDD argument, nope I cannot buy an SSD NAS or DAS or even multiple high TB drives, an 8TB SSD is insanely expensive so forget the 10/12TB multidrive setup with mirroring and with stupid QLC and no more MLC it's going to get worse. Last best MLC SSD is 860 Pro, it only has 4TB max. Shame that 2.5" are a dying breed now. NVMe's MLC is 970Pro at 1TB max.
Posted on Reply
#39
lexluthermiester
tiggerStill slower than an average SSD/m2 drive
And? Unless you want to spend $15,000, no SSD/M.2 is getting anywhere near 20TB. Mass storage doesn't need the blazing speed of SSD's, OS boot drives do.
Posted on Reply
#40
windwhirl
Ashtr1xReliability matters, I would like to see these come to market quick. WD with their superb HGST technology / Ultrastar vs Seagate whoever brings these first they win. As for SSD vs HDD argument, nope I cannot buy an SSD NAS or DAS or even multiple high TB drives, an 8TB SSD is insanely expensive so forget the 10/12TB multidrive setup with mirroring and with stupid QLC and no more MLC it's going to get worse. Last best MLC SSD is 860 Pro, it only has 4TB max. Shame that 2.5" are a dying breed now. NVMe's MLC is 970Pro at 1TB max.
True that. Just in case, I'm clarifying that I don't think the migration from HDDs to SSDs in enterprises will happen overnight. It's too expensive right now and the very limited write cycles of TLC and above just aren't up to the task in some scenarios.
Posted on Reply
#41
lexluthermiester
windwhirlI'm clarifying that I don't think the migration from HDDs to SSDs in enterprises will happen overnight.
True.
windwhirlIt's too expensive right now and the very limited write cycles of TLC and above just aren't up to the task in some scenarios.
Also true. Without a quantum leap in NAND durability, SSD's will never completely replace HDD's in the enterprise sector, the professional sector or even in certain consumer sectors.
Posted on Reply
#42
Rei
lexluthermiesterWithout a quantum leap in NAND durability, SSD's will never completely replace HDD's in the enterprise sector, the professional sector or even in certain consumer sectors.
Yeah, the solution to that would be either a breakthrough in making TLC & QLC be durable enough for longevity & reliability along with 3D NAND & V-NAND or go back to SLC & MLC. The latter option would need to have SLC & MLC to be in really high demand to have the price come down for a better, affordable, high capacity drives.

I think the enterprise sector & maybe the professional sector is still using eMLC.

Correct me if I'm wrong, but it seems like 3D NAND & V-NAND has been doing only decent job of complimenting TLC & QLC's longevity & reliability. Any takes?
Posted on Reply
#43
windwhirl
ReiYeah, the solution to that would be either a breakthrough in making TLC & QLC be durable enough for longevity & reliability along with 3D NAND 7 V-NAND or go back to SLC & MLC. The latter option would need to have SLC & MLC to be in really high demand to have the price come down for a better, affordable, high capacity drives.

I think the enterprise secotr & maybe the professional sector is still using eMLC.

Correct me if I'm wrong, but it seems like 3D NAND & V-NAND has been doing only decent job of complimenting TLC & QLC's longevity & reliability. Any takes?
There is this thing on Wikipedia:

But the sourcefor the table is almost a decade old, so there may have been some improvements.
Posted on Reply
#44
Rei
windwhirlThere is this thing on Wikipedia:

But the sourcefor the table is almost a decade old, so there may have been some improvements.
Good enough, I suppose... But then again, it doesn't look that old though. It has mention of Kioxia which was formed 1-2 years ago. Although, it could be just the table being updated but not straight from the source. Just as QLC is mentioned but no numbers on the table.
Anyway, thanks for this, Windwhirl.
Posted on Reply
#45
windwhirl
ReiGood enough, I suppose... But it doesn't look that old though. It has mention of Kioxia which was formed 1-2 years ago. Although, it could be just the table being updated but not straight from the source. Just as QLC is mentioned but no numbers on the table.
Anyway, thanks for this, Windwhirl.
The source itself for the endurance ratings is that article from 2012. The mention of Kioxia is probably someone doing some maintenance on the article (I did something similar on the Spanish side for some articles a long time ago, although on a completely different field, not related to technology).
Posted on Reply
#46
Unregistered
lexluthermiesterAnd? Unless you want to spend $15,000, no SSD/M.2 is getting anywhere near 20TB. Mass storage doesn't need the blazing speed of SSD's, OS boot drives do.
I have nothing but 2 m2's in my machine. Don't need a feck load of storage like some. If i need more, i will buy 2.5" ssds. i refuse to have noisy, slow mechanicals in my rig.
Posted on Edit | Reply
#47
lexluthermiester
tiggerI have nothing but 2 m2's in my machine. Don't need a feck load of storage like some. If i need more, i will buy 2.5" ssds.
Hey that's cool, if you don't need it you don't need it. Some do, like me.
tiggerIf i need more, i will buy 2.5" ssds. i refuse to have noisy, slow mechanicals in my rig.
Then it would seem that you haven't owned a HDD in a long time. Most HDD's made in the last 5 years are VERY quiet, even the 7200RPM variants.
Posted on Reply
#48
dragontamer5788
ReiYeah, the solution to that would be either a breakthrough in making TLC & QLC be durable enough for longevity & reliability along with 3D NAND 7 V-NAND or go back to SLC & MLC. The latter option would need to have SLC & MLC to be in really high demand to have the price come down for a better, affordable, high capacity drives.

I think the enterprise secotr & maybe the professional sector is still using eMLC.

Correct me if I'm wrong, but it seems like 3D NAND & V-NAND has been doing only decent job of complimenting TLC & QLC's longevity & reliability. Any takes?
3d NAND / V-NAND helps a bit, but not by a huge amount.

What's really helped is a whole slew of new controller algorithms that load-balance better. In particular, the proliferation of "static wear leveling", which means static blocks move around the drive.

For example: lets say you have a 350GB drive, and 300GBs of it is used "statically" (write once, read many use case. Like videos, pictures, archived tax documents, and the like). Under older algorithms (just 5 years ago), only the "free" 50GBs was wear leveled. (IE: Dynamic wear leveling). Under today's controller algorithms, all 350GBs are wear-leveled. Those "rarely written" video files will be moved to other blocks, so that those cells can be "cycled".

As such, a 1TB TLC SSD will have very good wear leveling, there's almost always a large amount of static-files in a typical consumer workload.
Posted on Reply
#49
windwhirl
lexluthermiesterThen it would seem that you haven't owned a HDD in a long time. Most HDD's made in the last 5 years are VERY quiet, even the 7200RPM variants.
Might depend on the user's perception of sound. Myself, I probably couldn't care less if the drives were rated at 30 db, since I live just two floors above an avenue very near to downtown Buenos Aires, so it's noisy pretty much all day until midnight or so. Though I do admit that I bothered to check the exact revision of the IronWolf drive I bought last month for, among other things, better acoustics (just 18 db for idle).
Posted on Reply
#50
Rei
tiggeri refuse to have noisy, slow mechanicals in my rig.
lexluthermiesterThen it would seem that you haven't owned a HDD in a long time. Most HDD's made in the last 5 years are VERY quiet, even the 7200RPM variants.
I have 3x external HDD, 3x 2.5" SATA HDD, 2x 2.5" PATA HDD & 2x 3.5 HDD & some range around a decade old while the rest is more than a decade old, yet none of them makes any noise beyond ambient noise even to this day.
Posted on Reply
Add your own comment
May 2nd, 2024 15:51 EDT change timezone

New Forum Posts

Popular Reviews

Controversial News Posts