Hard drive sizes have stopped improving

M

MarshMellow

Weaksauce
Joined
Oct 8, 2009
Messages
81
Anyone notice that we've had 2TB drives for 12? months now and there has been nothing new coming out or even murmurs of new capacities. What is up with this, when can we expect the 4TB drives and 1TB platters?
 
Last edited:
andy_rooney.jpg


(just imagine stacks of harddrives instead of the magazines)
 
Last edited:
I'm being serious here. Have we hit a limit on platter density? It's the most stagnant the traditional HD industry has been for at least a decade.
 
The only area traditional HD's compete well with SSD is size, so you would be thinking they would be increasing size as quick as possible.
 
There is no real consumer need for 4TB drives. Seriously there just isn't unless your downloading over 100 movies a month....
 
maleficarus said:
There is no real consumer need for 4TB drives. Seriously there just isn't unless your downloading over 100 movies a month....
Click to expand...

If you rip bluray movies then there is definitely a consumer need for more size. You could only fit ~ 38 or so on a 2TB hard drive.
 
I don't think it's been that long. The first 2TB drive, the WD20EADS, came out around April/May of 2009. So that was only about 9 months ago.

It was the first drive with 500GB platters I believe. And was only 5400 RPM. It would take a few more months before we'd even see the first 7200 RPM drive with 500GB platters.

Seagate and Samsungs 2TB drives are still relatively new. So I don't think there has been any slow down. 500GB platters at 7200 RPM haven't been around that long really.
 
I have to agree that it seems storage capacities have plateaued. I'm waiting for 2TB drives to drop to about $100, and I'm not sure when that'll be. From what I remember, manufacturers have reached something of a brick wall in terms of density, so the only way to add capacity is to add platters. Of course, I hope innovations continue!
 
The first 500gb platter drive was released ~Jan 2009, however they had been releasing info on the upcoming releases for something like 6 months (at least). We still haven't heard anything about 640/750GB platters which means they are at least 6 months away from an update.
 
Furthermore, 320GB platters were released by Western Digital in March 2008, less than a year later the 500GB platters were out. We are now more than a year later than the 500GB platters and there hasn't even been an announcement, signaling there has been a slow down.
 
With the 25nm NAND coming out from IMFT later this year, it should be possible to put 2 TB of NAND in a 3.5" form factor (the Colossus already has a 1 TB 3.5" SSD). So SSDs could catch up with HDDs in capacity this year. Of course, a 2 TB SSD would cost a fortune, but time will eventually solve that problem.
 
What is the inherent cost in flash manufacturing? And when can we expect the price to come down, when they get smaller than 25nm? Will 25nm 1TB drives *ever* be cheap?
 
MarshMellow said:
What is the inherent cost in flash manufacturing? And when can we expect the price to come down, when they get smaller than 25nm? Will 25nm 1TB drives *ever* be cheap?
Click to expand...

Why ask why?

Seriously, none of the questions you're asking has a concrete guaranteed answer.

Wait and see like everyone else.
 
The sooner Andy Rooney tackles the platter density crisis, the sooner we get JUSTICE.
 
Last edited:
I hope they dont rush into these things. It seems that the higher density the drive, the less reliable they've become. Of course I dont have a large enough sample size to make this conclusion...but it seems that a good portion of my newer hard drives develop bad sectors or write errors within a few years. Never seen this problem with 10 year old drives that may have had 50K+ operating hours.
 
MarshMellow said:
Furthermore, 320GB platters were released by Western Digital in March 2008, less than a year later the 500GB platters were out. We are now more than a year later than the 500GB platters and there hasn't even been an announcement, signaling there has been a slow down.
Click to expand...

Each progressive step in terms of density requires more effort then the step before. The fact that we haven't seen >500GB platters is not surprising. The HDD manufacturers are going to publically respond to the market. Right now the market is asking for cheap SSD's. I imagine the manufacturers have some R&D people squirreled away in a back room somewhere working on higher density platters.
 
Hard-drive capacity is growing slower than before, but is it worse now than a little while ago? I seem to remeber that 1 TB was the largest size a pretty long time too.
 
I read an article yesterday that since the introduction of 1TB drives traditional RAID5 has lost it protective capabilities because of the density of a single drive. A 7 disk RAID5 array would have a 20% chance of failure in the first year. RAID6 will reach the same problem soon once we'll see >2TB drives
There was no maths to back up the claim but I found it an interesting statement.
 
....? i don't understand how one drive failure of a 1TB drive in a 7disc raid5 array would be less protective than a 500gb disc in a 7 disc raid5 array
 
Maybe the MTBF (mean time between failures) has (significantly) decreased for larger drives?
 
There were just posts the other day in one of the RAID threads from a storage engineer who had worked for Hitachi in the past year. He said that many of the HDD manufacturers planned on a 4TB drive by Q4 of this year... Obviously this is second or third hand at best, but the guy seemed to know what he was talking about. I don't think manufacturer's plan on stopping at 2TB, but maybe the step from 2 to 4 is harder than it was from 1 to 2 (or it's that and a combination of less overall consumer demand for 4TB drives). While a lot of people on this forum probably can't wait to get their hands on a 4TB drive and build a monstrous array from it, I'm guessing Joe Public probably can't even fathom what to do with 4TB right now. (give him a few years though, he'll catch up) ;)
 
MarshMellow said:
If you rip bluray movies then there is definitely a consumer need for more size. You could only fit ~ 38 or so on a 2TB hard drive.
Click to expand...

You can fit about 90+ if you rip them to MKV + AAC audio. That's what I do with mine. For example, The Hangover is only abot 13 gigabytes. No loss in quality.

Even Akira with it's 13.5GB Japanese audio track :)eek:) came down to about 16 gb total after converting that 13.5GB 192 khz 24bit 5.1 DTS-HD MA track to a 900mb to a 96 khz 5.1 (32 bit) AAC track. No perceptual loss in quality.

Most blu rays only have about 18 GB of useful data if you only care about the movie and aren't anal about needing lossless audio. Copy video with no transcode.
 
Well we have kinda hit a platter density wall with the current tech.

4k Blocks should fix this and allow the HDD manufacturers to get larger disk out in the future.
Currently WD is the only mgfr that i know of that has a 4k block drive...WD20EARS.
So i expect WD to be the first to market with a larger capacity.

two years ago when the 1TB Hitachis hit the market. Hitachi said in a press release that we would have 5TB disk in 2010.
The clock is ticking Hitachi :)
 
BENN0 said:
Maybe the MTBF (mean time between failures) has (significantly) decreased for larger drives?
Click to expand...

That should decrease reliability over the board, not just for raid, in fact, one can argue that raid is more useful if drives brake more often.
 
I know I've seen drive failure rate go through the roof. It seems low capacity drives used to never die. I have large capacity drive die much too often.
 
If you rip bluray movies then there is definitely a consumer need for more size. You could only fit ~ 38 or so on a 2TB hard drive.
Click to expand...
If you have the money to buy that many Blu-rays, I think it would be perfectly within your capability to buy another hard drive.
 
robble said:
I know I've seen drive failure rate go through the roof. It seems low capacity drives used to never die. I have large capacity drive die much too often.
Click to expand...

More points of failure is what it really comes down to.

However if you look at failures per GB its actually lower than it used to be.

Let me explain.
Before when there was 300gb drives you only had 300gb of space that had the potential for messing up, bad blocks, URE, etc.
Now you have drives that have 6x the capacity so the chance of the drive failing is higher because there is more opportunity for a messup.

But if you look at it proportionally you would have to use 6x 300gb drives to get about the same space as a 2TB.

with 6x300gb drives your MTBF is going to be much less than with 1x 2TB.

So in reality is actually better than it used to be per GB its just noticed more often because people are using less spindles.
 
nitrobass24 said:
More points of failure is what it really comes down to.

However if you look at failures per GB its actually lower than it used to be.

Let me explain.
Before when there was 300gb drives you only had 300gb of space that had the potential for messing up, bad blocks, URE, etc.
Now you have drives that have 6x the capacity so the chance of the drive failing is higher because there is more opportunity for a messup.

But if you look at it proportionally you would have to use 6x 300gb drives to get about the same space as a 2TB.

with 6x300gb drives your MTBF is going to be much less than with 1x 2TB.

So in reality is actually better than it used to be per GB its just noticed more often because people are using less spindles.
Click to expand...

That makes sense unless you realize it's not usually the platters themselves that are messing up but thing like armatures and circuit boards in the drives that are dying. When drives make clicking noises it's not a result of capacity.
 
robble said:
That makes sense unless you realize it's not usually the platters themselves that are messing up but thing like armatures and circuit boards in the drives that are dying. When drives make clicking noises it's not a result of capacity.
Click to expand...

Thats kind of my point.
Your chances of it not happening are better with larger capacity drives because there are less armatures and circuit boards to fail.

A group of 6 drives have a better chance of a single failure, vs a single drive.
 
on top of the 500GB platters, WD now has the EARS drives, with the 4k sectors, which were said to increase the density by 20-30%?meaning a 2.5TB drive should be coming out soon :p

Also I thought the Hitachi 2TB drive is a 5platter drive, why not stick 5 500GB platters in there for a 2.5TB I think another problem is formatting said drive, doesn't NTFS have an issue with >2TB?
 
I didn't realize there was a 5 platter 2tb drive. Interesting.
In that case there are already some 640GB per platter drives on the market, why not do 640GBx5 and get a 3.2TB monster drive? maybe we will see that soon too.
 
Here I sit loading windows onto a new wd 1tb black hd because my seagate 7200.11 1tb decided to die. =(


It may not be a bad thing to not get drives that are 5tb. Imagine all the porn you could lose.

Lucky I had recently built this computer (reused the drive) and there wasn't a lot of stuff on there that can't be downloaded from steam again. I do hate losing a bunch of emails/programs/setup though. It's such a pain to set back up.
 
True, but I probably wouldn't install my OS on a 5TB drive if I could avoid it. :) It's a lot easier to have a small boot drive that you can image regularly so that if it crashes you can restore the complete image and go on your way. Sorry about the HDD crash, it's been about a year since my last one (non boot drive) and I feel your pain.
 
nitrobass24 said:
More points of failure is what it really comes down to.

However if you look at failures per GB its actually lower than it used to be.

Let me explain.
Before when there was 300gb drives you only had 300gb of space that had the potential for messing up, bad blocks, URE, etc.
Now you have drives that have 6x the capacity so the chance of the drive failing is higher because there is more opportunity for a messup.

But if you look at it proportionally you would have to use 6x 300gb drives to get about the same space as a 2TB.

with 6x300gb drives your MTBF is going to be much less than with 1x 2TB.

So in reality is actually better than it used to be per GB its just noticed more often because people are using less spindles.
Click to expand...


Well said. Also, I think possibly manufacturers are kind of dragging their feet for a few reasons. 1) SSD's are the hot item right now 2) As drive sizes increase, theres a chance ( debatable, as weve seen it before) that your raid 5 and 6 becomes less effective , because during a rebuild, you have a higher chance of data corruption due to the huge sizes of pools and long times to rebuild arrays ( and I think some raid cards/setups actually allow for this with some sort of protection) 3) The technology just isnt quite up to par to announce anything 4) maybe waiting for the next Sata platform ?? Makes sense to me, what good is 40 gazillion gigs of storage if it takes a week to access it.

Anyway, just some of my thoughts about it.
 
You must log in or register to reply here.
Back
Top