News Seagate: We Are On Track with 20TB HAMR HDDs in December

We need to mark our calenders down for Dec 2026 and see if Seagate kept their promises
Long-term hard drive timelines rarely pan out. Look at this Tom's article about Seagate's HAMR from 2010. It sounds like this was just speculation on the writer's part based on the suggested density increase, but they wrote...
With roughly 50x the potential of PMR, HAMR should lead the way beyond 100 TB drives and possibly into the region of 200 – 300 TB in the 2020 to 2025 time frame.
https://www.tomshardware.com/uk/news/hdd-ssd-harddrive,11048.html

Now, we're just getting 20TB, and are only forecasting 50TB by 2026. This Seagate press release from 2013 did suggest "20TB by 2020" which is a lot more accurate, though it sounds like they were intending that to be a 2.5" 10K drive, while this is probably going to be a 3.5" drive at a slower rotational speed...

And this article from 2017 suggested that Seagate would have lower-capacity HAMR drives on the market by 2018, and that "the company has plans to bring 20TB+ drives to market in 2019"...
https://www.tomshardware.com/news/seagate-wd-hamr-mamr-20tb,35821.html
 

nings

Commendable
Mar 5, 2018
14
2
1,515
0
Still wait my 1po drive !

Hdd evolve really fast from some mb at 4gb to 1to was very fast. You never know how to fullfil these drive each time you buy a new hdd. Since 1to is out all has slowing down. We should already be at 1po per drive if continue same progress than before. I have already 22to and it's almost full need a lot more than that 100/200to for have the same space than when I switch from some mb to 4gb drive.
 
Oct 28, 2020
3
0
10
0
Is it just me or does the math on hard drives just not add up to anyone else? I have been thinking about this for several years and I must be missing something. Lets just assume for simplicity that each platter in a 3.5" HDD has a radius of 1", in reality it is problably close to 1.5", but I'm just going for simple numbers here. With a radius of 1" you have a surface area of a little over 3 sq in. per platter face, multiply that by 2 for a platter surface area of 6 square inches. Multiply that by the 9 platters, the typical amount in leading high capacity disks and you get 54 sq inches of surface area. Leading edge tech is getting about 1TB per sq inch now so that leaves me figuring 54 TB in a drive available today. Now I know you couldn't use every bit of area perfectly, but I find it strange drives are only approaching a third that capacity and I estimated on the low side. Where is my logic breaking down?
 

TMTOWTSAC

Honorable
Jun 27, 2015
1,461
4
11,965
225
Areal density is usually given in bits, not bytes, so you'd need to divide by 8 for the units to match up. 54 terabits would only equal 6.75 terabytes. Now to line up with modern drives you'd need to reclaim that platter size from your estimate, and use the latest areal density numbers. Not sure exactly where we're at now, Seagate released a press statement in 2015 claiming 1.34 Tbit/sq inch, but that was almost certainly using SMR.
 

nofanneeded

Respectable
Sep 29, 2019
1,562
251
2,090
52
Is it just me or does the math on hard drives just not add up to anyone else? I have been thinking about this for several years and I must be missing something. Lets just assume for simplicity that each platter in a 3.5" HDD has a radius of 1", in reality it is problably close to 1.5", but I'm just going for simple numbers here. With a radius of 1" you have a surface area of a little over 3 sq in. per platter face, multiply that by 2 for a platter surface area of 6 square inches. Multiply that by the 9 platters, the typical amount in leading high capacity disks and you get 54 sq inches of surface area. Leading edge tech is getting about 1TB per sq inch now so that leaves me figuring 54 TB in a drive available today. Now I know you couldn't use every bit of area perfectly, but I find it strange drives are only approaching a third that capacity and I estimated on the low side. Where is my logic breaking down?
Simple answer : They wont give you the best they have until they have another technology .

That is , every two years they increase the capacity and take your money while claiming : new barrier , while it is the same old technology.

Look at intel staying at 4 cores max for ten years while they had the technology to release upto 10 cores from day one ...

Monopolies know what they will sell you ten years in advance at least in their road maps and they know it exactly from day one . they know the capacity of their technology but they will give it to you step by step ..
 

spongiemaster

Respectable
Dec 12, 2019
1,398
660
2,060
0
Leading edge tech is getting about 1TB per sq inch now so that leaves me figuring 54 TB in a drive available today. Now I know you couldn't use every bit of area perfectly, but I find it strange drives are only approaching a third that capacity and I estimated on the low side. Where is my logic breaking down?
Seagate says the areal density of their new HAMR drives is 2 Tb/sq in. That's bits, not bytes. So about 250GB/sq in. 3.5" drive platters have an actual diameter of about 3.75". Then you have to remove the center spindle, and you'll end up pretty close to the 20TB drives they're announcing here.
 

ASK THE COMMUNITY