Tom's Guide > Forum > Storage > Hard Disks > 1TB milestone reached

1TB milestone reached

Forum Storage : Hard Disks - 1TB milestone reached

TomsGuide.com: Over 800,000 questions and answers to address all your high-tech questions. Sign up now! Its free!
Page:    Previous 1 2 Next Bottom Search this thread
Word :    Username :           
 

Well... Seagate is apparently going to show off their 1TB hard drive next week at CES.

http://www.tgdaily.com/2007/01/04/seagate_1tb_drive/

Amazing how far storage has come in just 5 years.

Sponsored Links
Register or log in to remove.
- 0 +

thats pretty awesome however im more worried about how the thing will perform i still think even tho with all the improvements going on that hdds have a long way to go

Reply to yakyb

Space wise they have achieved quite something in recent years. Speed wise, well, things are standing still. Makes me wonder about the HDD industry.

Reply to Slobogob
- 0 +

It will perform better, how much no one knows.
The 750GB seagate drive used 5 platters.
This new 1TB drive uses 4.

This means Density is a bit greater which means faster reads and writes.
I am eagerly waiting for this guy.

Reply to Kurz

Score! Now I can replace all 5 of my hard drives with just one - and then pray that it won't fail!

Honestly though, (and this is in no way an 'OMFG Mac OS is ttly 1337' comment in case anyone wants to pounce on me) I really don't trust Windows not to f* something up with a drive that large - it's just the sort of thing that would be buggy in Windows. I won't go higher in capacity than my 320gb - it's just too risky.

Reply to mesarectifier
- 0 +

Quote :

Vertical recording technology, and flash HDD caches
should speed things up quite a bit:

http://www.tgdaily.com/2007/01/04/ [...] _32gb_ssd/


Sincerely yours,
/s/ Paul Andrew Mitchell
Webmaster, Supreme Law Library
http://www.supremelaw.org/



vertical recording is actually slower than the standard non-vertical drives, it just allows more data density (bigger capacity) is all.

Reply to niz

In 1980 the ibm hdd for their 3380 had a seek time of 16ms if i recall correctly. Now, almost three decades later we´re at what? 4.5?

I´ve to admit that space and size have changed quite a lot, but speed doesn´t seem to improve at all (heck, maybe a single ms every odd year!!!). My point is, i don´t need 1 TB. I don´t need even half of that. Not even 250GB. A little more than just 100GB would suffice and i´m certain that a few GBs are more than enough for the average user. Still, everyone waits for the computer to boot or to copy a file from a to b. Why is that?
I mean, it took more than fourty years to end the reign of punch cards (in computing, i believe they´ve been used in telegraphy before) and another ten to get rid of magnetic tape. Last year the HDD concept had it´s 50th anniversary. That´s pretty scary.
The way data is stored within a PC is overdue for some major changes. I mean, hdds have evolved, sure. But at the speed machines evolve at, not computers.
I really, really hope the SSD goes mainstream (and becomes cheaper!).

Reply to Slobogob
- 0 +

1 Tb, that's a lot of porn. 8)

Reply to MrMr
- 0 +

To Niz saying "Vertical recording is slower"
You are sadly mistaken.

Also what you mean is Perpendicular recording.

Reply to Kurz

Quote :

1 Tb, that's a lot of porn. 8)



I already can see the ads:

"Store your favorite photos, movies, music and files on a Seagate Barracuda 1 Terabyte harddrive! Seagate - Store the content of your life.


Buy now and get a tube of Astroglide for free!"

8O

You got me worried there.

Reply to Slobogob
- 0 +

Quote :

Space wise they have achieved quite something in recent years. Speed wise, well, things are standing still. Makes me wonder about the HDD industry.


yeh, pretty pathetic. The next obvious evolution to hard memory is solid state (read "flash" ) memory. but needs to be 40 to 100 G to serve the Vista, the killer whale.

- need to get faster access, read, and write
- need to get away from the risks that mechanical devices bring along with them

Reply to Pepicek

Its easier to increase size in big increments (since, comparatively speaking, the amount of space on disk platters is huge compared with the size of the magnetic granules, small increase in magnetic density reap huge wins), its much harder to decrease access time and the decreases come in much smaller increments because fundamentally you are fighting a mechanical process. Think about CD->DVD->BR/HDDVD - the storage rates have gone up by a huge factor but it still takes a butt load of time to move the head from one edge of the disk to the other, settle it, focus the lens (not sure what the equivalent is for hard drives but I'm sure there is one), locate the heads actual position on the disk and then wait for the correct data to come by.

Its possible that the only way we can acheive better access times (I mean in the order of one or two magnitudes rather than a ms here or there) is to go with completely different technology like holographic memory that have no phyiscially moving parts.

Reply to Bozzkins

Quote :


Its possible that the only way we can acheive better access times (I mean in the order of one or two magnitudes rather than a ms here or there) is to go with completely different technology like holographic memory that have no phyiscially moving parts.



That´s exactly what i am thinking. The mechanical HDDs are overdue to become obsolete. I´m still curious to see what will happen. I think SSDs are going to be "it", but that would mean the mass storage market leadership would change hands from Seagate to Samsung. While thats a possibility, i don´t see that happening without a really, really tough fight.

Reply to Slobogob

oh well we shall see.

Reply to BuddyGoodness

Chances are most of us hardcore types will have 30 - 60 gig nand flash storage by christmas.

Reply to jkflipflop98

Quote :

Its possible that the only way we can acheive better access times (I mean in the order of one or two magnitudes rather than a ms here or there) is to go with completely different technology like holographic memory that have no phyiscially moving parts.



Well they could do that, and the performance may be great. But, and it is a very big but, at first it will be very expensive, and it will take very long to be out in the market in full force (meaning cheaper and has enough density), probably 15-20 years or more.

Like what Seagate states about full flash hard drives, they don't see the need to shift to manufacturing flash drives, as simply the cost of magnetic hard drives is way smaller than manufacturing flash based hard drives.

Reply to amnotanoobie

I suppose the other half of us hardcore types will be in jail after trying to rob a bank to get their SSD fix. :?

Reply to Slobogob

Quote :


Well they could do that, and the performance may be great. But, and it is a very big but, at first it will be very expensive, and it will take very long to be out in the market in full force (meaning cheaper and has enough density), probably 15-20 years or more.

Like what Seagate states about full flash hard drives, they don't see the need to shift to manufacturing flash drives, as simply the cost of magnetic hard drives is way smaller than manufacturing flash based hard drives.


That´s exactly the attitude that forced IBM out of the desktop market, that made Bill Gates make his infamous memory statement and that will ruin Seagate if they don´t watch out.
Just look at what flash memory cards did on the market already. They did what Zip-drives and lots of other in-between devices failed to do - they made the floppy (almost) obsolete. Then take a look at the prices of 1 or 2 GB memory sticks and compare them with the prices of the same sticks last year. The market for NAND memory isn´t getting smaller. Intel, Micron and Samsung are expanding their fabs and competing for speed and size.

Look at these prices and then tell me again that you really believe it´s going to be more than three or four years until SSDs are widely available?

Reply to Slobogob
- 0 +

Flash will not be the answer to our prayers.
Flash still has a huge handicap, which is the finite amout of rewrites it can perform per bit is about 100,000. Another thing is it access via data banks instead of other new potential memory technologies.

I will probably never get a flash drive for those very reasons.

PRAM and NRAM will be the memory of choice in the future.
Wiki them.

http://en.wikipedia.org/wiki/Phase-change_memory

http://en.wikipedia.org/wiki/Nano-RAM

Either of those I would be happy with.

Reply to Kurz
- 0 +

Quote :

1 Tb, that's a lot of porn. 8)



I already can see the ads:

"Store your favorite photos, movies, music and files on a Seagate Barracuda 1 Terabyte harddrive! Seagate - Store the content of your life.


Buy now and get a tube of Astroglide for free!"

8O

You got me worried there.I guess you guys don't remember this? :P

Quote :

"Let's face it, we're not changing the world. We're building a product that helps people buy more crap - and watch porn."



http://money.cnn.com/2006/11/30/ma [...] /index.htm

Reply to 1Tanker
- 0 +

Quote :

Well... Seagate is apparently going to show off their 1TB hard drive next week at CES.

http://www.tgdaily.com/2007/01/04/seagate_1tb_drive/

Amazing how far storage has come in just 5 years.

Not sure, but it looks like Hitachi announced theirs first.

http://www.hitachigst.com/portal/s [...] eTok=token

They're quoting a $399USD price, whereas Seagate hinted at ~$700. :o Though the Hitachi utilizes 5 heads vs. the Seagate's 4, the Hitachi has a 32MB buffer. :lol: Can't wait to see these beasts in a shoot-out.

Reply to 1Tanker
- 0 +

Quote :

I will probably never get a flash drive for those very reasons.



You don't have a flash drive?

Sorry, I'm having a hard time wrapping my mind around this one.

When I made a year at work they made quite the show of giving me a 2G drive because I'm internally famous for walking up, plugging in my "dongle", and fixing systems. Later, as I was adding it to the keychain I keep them on the boss asked what I was doing. I told him what I was doing, adding it to the 2-1G's and 2-4G's I already had. He took it back. I now have an 8G from the company. Kripes, I got 3 128's and 2 64's this year from sales reps. Loaded them up with MP3's and regifted. My *kids* have 1Gers.

Nope. Still can't get it. Not gonna try anymore.

Reply to MISRy
- 0 +

Quote :

Flash will not be the answer to our prayers.
Flash still has a huge handicap, which is the finite amout of rewrites it can perform per bit is about 100,000. Another thing is it access via data banks instead of other new potential memory technologies.

I will probably never get a flash drive for those very reasons.

PRAM and NRAM will be the memory of choice in the future.
Wiki them.

http://en.wikipedia.org/wiki/Phase-change_memory

http://en.wikipedia.org/wiki/Nano-RAM

Either of those I would be happy with.

I guess you don't want a digital camera either? :roll:

Reply to 1Tanker

Quote :

Flash will not be the answer to our prayers.
Flash still has a huge handicap, which is the finite amout of rewrites it can perform per bit is about 100,000. Another thing is it access via data banks instead of other new potential memory technologies.

I will probably never get a flash drive for those very reasons.


The rewrite-problem bothered me too. Thats why i won´t buy any of those hybrid drives. I think an external Stick will do fine to speed up boot times. The thought of buying a hd and the Flash part breaking down after 2 years while the rest of the HD is still good for another five scares me off.
Yet i´m certain they can and will improve the limit. The question is how far can they improve it and will it be fast enough before another technology emerges.

Quote :

PRAM and NRAM will be the memory of choice in the future.
Wiki them.

http://en.wikipedia.org/wiki/Phase-change_memory

http://en.wikipedia.org/wiki/Nano-RAM

Either of those I would be happy with.



Nice read. I read about those nano-tubes before. As they already mentioned, they have to get production to a level that can compete with flash. And that will take quite some time. Makes me think though. Hopefully flash is just a intermediate solution then.

Reply to Slobogob

Quote :

Flash still has a huge handicap, which is the finite amout of rewrites it can perform per bit is about 100,000.



I don't believe that's a problem.

The controllers on all modern flash memories do write-leveling, where each write to the device is evenly distributed across all the flash memory available. In other words, if you constantly load a file, update it, and save it, the file isn't saved to the same block of flash memory every time. The write hardware ensures that it is saved to whichever block of flash has currently been written to the least. Thus, over time, the write logic evens the wear on the entire device, making it so that no one block of memory approaches 100,000 writes before others.

To see the margin to failure that this gives you, lets do some quick calculations for a heavy-use hard drive. Let's assume the following:
[*:32b4258f6a]64MB of pagefile updates every 1 minute = 184,320,000 blocks updated per day (block = 0.5K).
[*:32b4258f6a]5MB of registry updates every hour = 240,000 blocks updated per day.
[*:32b4258f6a]10MB of files updated every 15 minutes = 1,920,000 blocks updated per day.
[*:32b4258f6a]20MB of Windows updates, updated every week = 5714 blocks updated per day.

Total 512-byte blocks updated every day (for 24-hour/day operation) = 186,485,714.

For a 32GB flash device, the device has 62,500,000 512-byte blocks, each of which can handle 100,000 writes = 6.25 x 10^12 writes available.

6.25E12/186485714 = 33,514 days, or 92 years until a block reaches 100,000 writes and fails.

That's for fairly heavy hard drive use. If the computer has enough RAM, the pagefile updates go away, extending the life significantly.

Furthermore, most flash devices have on board spare blocks. For example, if the device is a 32GB device, it will report it's size to the computer as a 31.95GB device, and keep 50MB of blocks in reserve. If a block fails, the device will allocate a spare block from that pool. When the pool of spare blocks goes below a certain threshhold, it will generate a SMART error to the computer, probably more reliably than hard drives do.

I wouldn't worry too much about the limited writes on flash causing problems.

Reply to SomeJoe7777
- 0 +

You've made a valid point. I completely forgot about the write-leveling feature. Still though flash is still slow compared to other Tech's over the horizon. It can still act as a go between between the techs.

Though I probably never pay for one of these drives till the cost comes down from ~$15 per gigabyte to somewhere around 5 or less. That'll only happen with smaller processes.

To those who wavered from the discussion to the portable flash drives. I wasnt talking about them. In fact I fully imbrace Flash and the prospect of Solid state drives is a great thing to have.

Waits for a bigger flash drive and smaller processes to drive costs down.

Reply to Kurz

Yeah... and now Hitachi is releasing one too! So seriously... how many of you guys are going to buy one? I know I'm going to buy at least 3.

Reply to Dante_Jose_Cuervo

Well my media server uses 2 500gb in raid 1. I will use this new technology to buy more cheaper 500gb drives as that seems to be right amount per drive.

100gigs per category stored is about right for the average downloader.

I.E 100gigs of music
100gigs of video (nondvd)
100gigs of TV
100gigs of games
100gigs of random files.


Then back it all up using raid (ya i know not really a backup solution). I need a 500gb drive to back up my 500(x2) raid 1 array.

---------------------
I was impressed with the first 1gb drives, the 20gb drives, the 40,80,160,.... then size really didnt make a great impression any longer. The 1tb is fabulous, but I dont consider it as big as a break through as the 1gb was.

I dont think I will be impressed again until I see 100tb drive, or 1000tb drive.

I would rather see Holographic technology become main stream, rather then seeing a 10/100tb drive. I have been following the technology for 10 years, and I am pleased to see the first disks coming out.

Reply to CompTIA_Rep
- 0 +

Quote :

Well my media server uses 2 500gb in raid 1. I will use this new technology to buy more cheaper 500gb drives as that seems to be right amount per drive.

100gigs per category stored is about right for the average downloader.

I.E 100gigs of music
100gigs of video (nondvd)
100gigs of TV
100gigs of games
100gigs of random files.


Then back it all up using raid (ya i know not really a backup solution). I need a 500gb drive to back up my 500(x2) raid 1 array.

---------------------
I was impressed with the first 1gb drives, the 20gb drives, the 40,80,160,.... then size really didnt make a great impression any longer. The 1tb is fabulous, but I dont consider it as big as a break though as the 1gb was.

I dont think I will be impressed again until I see 100tb drive, or 1000tb drive.

I would rather see Holographic technology become main stream, rather then seeing a 10/100tb drive. I have been following the technology for 10 years, and I am pleased to see the first disks coming out.

Seeing how it's not wise to stuff your HD to 100% capacity, 500GB won't do. :wink:

Quote :

or 1000tb.

= Petabyte. :)

Reply to 1Tanker

I know... I am just saying that 100gb per category is fine... and I could have named many more then 5, and most people wouldnt be collection 100gb of those.



and ya 1000=petabyte.

But I like long drawn out numbers.... like 1000 million (billion)

Reply to CompTIA_Rep

Wait... isn't it actually 1024TB in a petabyte? I mean I know about the whole manufacturer-vs-OS storage confusing but ya.

Reply to Dante_Jose_Cuervo

depends who you ask. If you ask Seagate.. there are 1000gb in a PB. If you ask Windows, therea re 1024.

Reply to CompTIA_Rep
- 0 +

Common users dont need 1TB, only people who work with servers and workstations.

I only have 1 500GB HDD for storage and thats it. My question is, why doesnt Seagate, Hitachi and Western Digital make faster hard drives instead of bigger hard drives? I would always prefer a 250Gb HDD with 10,000 RPM and 16mb of cache rather than a 1TB with 7200rpm and 32mb of cache.

Reply to slim142

Quote :

Common users dont need 1TB, only people who work with servers and workstations.

I only have 1 500GB HDD for storage and thats it. My question is, why doesnt Seagate, Hitachi and Western Digital make faster hard drives instead of bigger hard drives? I would always prefer a 250Gb HDD with 10,000 RPM and 16mb of cache rather than a 1TB with 7200rpm and 32mb of cache.



This raises a good point.

Why aren't they faster? Why don't they use higher rpms? Well... physics. I'm not going to go too far into detail but the reason SCSI, SAS and raptors can move so fast are because they use smaller platters. If I'm correct, raptors only use 2 platters (although I could be wrong, please correct me if the info is bad). The smaller... and sometimes fewer platters allow the motors to move faster to lower latency and whatnot... thus increasing throughput.

EDIT!: I understand that most people don't even need 1TB, but... I am one of those few people who would benefit from it. I ONLY use .wav files for my music... so that right there takes up tons of space (I have a band... that records... and we like the higher fidelity formats), that... and I have TONS of anime... most of it consists of fansubs... but that's besides the point. So... I've decided to get 6!(not 6 factorial... just six) of these TB hdds (and yes... I would actually be able to fill quite a bit of that up) and... put them in RAID 5...

Reply to Dante_Jose_Cuervo
- 0 +

Quote :

Common users dont need 1TB, only people who work with servers and workstations.

I only have 1 500GB HDD for storage and thats it. My question is, why doesnt Seagate, Hitachi and Western Digital make faster hard drives instead of bigger hard drives? I would always prefer a 250Gb HDD with 10,000 RPM and 16mb of cache rather than a 1TB with 7200rpm and 32mb of cache.



This raises a good point.

Why aren't they faster? Why don't they use higher rpms? Well... physics. I'm not going to go too far into detail but the reason SCSI, SAS and raptors can move so fast are because they use smaller platters. If I'm correct, raptors only use 2 platters (although I could be wrong, please correct me if the info is bad). The smaller... and sometimes fewer platters allow the motors to move faster to lower latency and whatnot... thus increasing throughput.

EDIT!: I understand that most people don't even need 1TB, but... I am one of those few people who would benefit from it. I ONLY use .wav files for my music... so that right there takes up tons of space (I have a band... that records... and we like the higher fidelity formats), that... and I have TONS of anime... most of it consists of fansubs... but that's besides the point. So... I've decided to get 6!(not 6 factorial... just six) of these TB hdds (and yes... I would actually be able to fill quite a bit of that up) and... put them in RAID 5...

Well if what you say is true, then why dont they start doing that? Ok Hitachi achived a milestone, interviews here interviews there so what? customers are not impressed and you can see it here. I have never read that Seagate or Hitachi are planning to release 10,000rpm hard drives. Not even a single rumor. Dont tell me "Cheetah" cuz that thing is expensive as hell. Not even a desktop user hdd.

BTW, Im not trying to say that "we dont need 1tb hdd", is just that MOST people dont need them. Is OK for the ones who do need them. Im not against the 1TB hdds, Im just complaining, they care so much in bigger drives, well why dont companies start creating faster drives too? for the other side where the users who need speed are!

Reply to slim142

You bring up a great point, and honestly, I have no idea why they don't make them faster... and ya... I don't like cheetahs... I've had too many of them die on me. I prefer Fujitsu's SCSI and SAS drives... much more reliable... and cheaper too!

And I agree with you... not many people really need a 1TB hdd... but then again... not everyone needs an X6800 or a QX6700 or an 8800GTX either!

Reply to Dante_Jose_Cuervo
- 0 +

Quote :

You bring up a great point, and honestly, I have no idea why they don't make them faster... and ya... I don't like cheetahs... I've had too many of them die on me. I prefer Fujitsu's SCSI and SAS drives... much more reliable... and cheaper too!

And I agree with you... not many people really need a 1TB hdd... but then again... not everyone needs an X6800 or a QX6700 or an 8800GTX either!



Well thats true, but graphics cards are updated rapidly. And even processors. On july we got Conroe, by november we had kentsfield already. In a couple of months after R600 is released we will have a war of Dx10 ready cards.
But that doesnt happen with hdds. 750gb was released almost a year ago! after 10 months approx we get 1TB. 10 months for extra 250gb? while videocards get new technologies and faster clocks every month?

I just dont understand that. but what can we do, we just have to wait sitting on our chairs to see what the companies do. thats reality.

Reply to slim142
- 0 +

Quote :

BTW, Im not trying to say that "we dont need 1tb hdd", is just that MOST people dont need them. Is OK for the ones who do need them. Im not against the 1TB hdds, Im just complaining, they care so much in bigger drives, well why dont companies start creating faster drives too? for the other side where the users who need speed are!

One good thing about ever-increasing HD sizes, is that it drives down the price of "normal" drives(i.e. 250GB/320GB). Being able to get 250GB for under $100 is nice. :)

Reply to 1Tanker
- 0 +

Not so easy for me.
1TB is big, but that is a whole lot of bits!
That means it can be easier for 1 bit to fail just because there are more bits.
For now I will stick with 400GB drives in RAID 5.

Actually, the 1TB drive has 4 platters. Why not have each platter spin seperately and have RAID 5 internally? Can't be THAT hard, 1" drives already exist,

Reply to enewmen
- 0 +

Quote :

Not so easy for me.
1TB is big, but that is a whole lot of bits!
That means it can be easier for 1 bit to fail just because there are more bits.
For now I will stick with 400GB drives in RAID 5.

Actually, the 1TB drive has 4 platters. Why not have each platter spin seperately and have RAID 5 internally? Can't be THAT hard, 1" drives already exist,

Actually, thats not a bad idea. The thing is, if your motor(bearings,etc.) goes..you still lose it all. :( I doubt they could use 4 seperate motors.

Reply to 1Tanker
- 0 +

Well thats the good part. Now, right now for $100 you can get a 7200.10 (Perp... Rec..) Barracuda of 320GB in newegg. I think they are at a good price right now so making their prices lower would just make no change for the people who still look for fast drives like the raptors.

Reply to slim142
- 0 +

Quote :

Not so easy for me.
1TB is big, but that is a whole lot of bits!
That means it can be easier for 1 bit to fail just because there are more bits.
For now I will stick with 400GB drives in RAID 5.

Actually, the 1TB drive has 4 platters. Why not have each platter spin seperately and have RAID 5 internally? Can't be THAT hard, 1" drives already exist,

Actually, thats not a bad idea. The thing is, if your motor(bearings,etc.) goes..you still lose it all. :( I doubt they could use 4 seperate motors.

The internal RAID system will be more reliable and redundant. Think of it as 4 completely seperate/enclosed super-slim drives in one 3 1/2" case. If one drive actually does fail, just replace the platter and the data gets recreated. Anyway, a lot more can be done.

Reply to enewmen

Yeah... I know what you mean. Lots of bits to fail... but that's why I still use RAID 5. Hotswap... as old as it is... is still a godsend... even on my desktop.

Reply to Dante_Jose_Cuervo

Quote :

Flash still has a huge handicap, which is the finite amout of rewrites it can perform per bit is about 100,000.



I don't believe that's a problem.

The controllers on all modern flash memories do write-leveling, where each write to the device is evenly distributed across all the flash memory available. In other words, if you constantly load a file, update it, and save it, the file isn't saved to the same block of flash memory every time. The write hardware ensures that it is saved to whichever block of flash has currently been written to the least. Thus, over time, the write logic evens the wear on the entire device, making it so that no one block of memory approaches 100,000 writes before others.

To see the margin to failure that this gives you, lets do some quick calculations for a heavy-use hard drive. Let's assume the following:
[*:5773e679eb]64MB of pagefile updates every 1 minute = 184,320,000 blocks updated per day (block = 0.5K).
[*:5773e679eb]5MB of registry updates every hour = 240,000 blocks updated per day.
[*:5773e679eb]10MB of files updated every 15 minutes = 1,920,000 blocks updated per day.
[*:5773e679eb]20MB of Windows updates, updated every week = 5714 blocks updated per day.

Total 512-byte blocks updated every day (for 24-hour/day operation) = 186,485,714.

For a 32GB flash device, the device has 62,500,000 512-byte blocks, each of which can handle 100,000 writes = 6.25 x 10^12 writes available.

6.25E12/186485714 = 33,514 days, or 92 years until a block reaches 100,000 writes and fails.

That's for fairly heavy hard drive use. If the computer has enough RAM, the pagefile updates go away, extending the life significantly.


That´s an excellent point! I didn´t think of that. Yet, there still is a problem.
The write-leveling can only work with the free parts of the flash drive. So filling a flash drive up to 90% would wear it out 10 times faster since the write logic could only use the remaining 10% for leveling. If i use the flash drive for a p2p programm now, you have a lot of writes within a very short time. The more you fill it up, the faster it breaks down. Okay, if i fill it up to the 90% i mentioned it´s still going to last quite a while (simply put 10% of what you calculated), which should still suffice.

Quote :


Furthermore, most flash devices have on board spare blocks. For example, if the device is a 32GB device, it will report it's size to the computer as a 31.95GB device, and keep 50MB of blocks in reserve. If a block fails, the device will allocate a spare block from that pool. When the pool of spare blocks goes below a certain threshhold, it will generate a SMART error to the computer, probably more reliably than hard drives do.

I wouldn't worry too much about the limited writes on flash causing problems.


Another good argument. With mechanical drives its just hit or miss. They either work or they break. At least in most cases. In that matter i´d really prefer a slowly fading Flash drive instead of a header crashed mechanical one. :lol:

Reply to Slobogob
- 0 +

My wife's small business has 1T (harddrive) of backups. I restore an old file about 2 times a year.

Most small businesses can use 1T of harddrive space. But 500GB drives are a lot cheaper than 1TB drives. Most of the workstations on small business networks gat by nicely on 40-80GB.

My PVR has 1.5T of harddrive space. Almost 500GB is filed by 200 old VHS tapes. 20 3 hour HDTV programs fill 500GB (DivX allows 60.) PVRs require lots of space - cheap space is better.

I don't think Harddrive speed is very important. For the most part my typing is too slow to drive a harddrive to its limits. Editing video is limited by CPU speed not Harddrive speed.

Reply to GeorgeH

1 word: fansubs. They take up a lot of space... with the amount I have. Honestly... 2TB would be great. All my .wav files and anime... *drools*.

Reply to Dante_Jose_Cuervo
- 0 +

Quote :

My question is, why doesnt Seagate, Hitachi and Western Digital make faster hard drives instead of bigger hard drives?



Because bigger hard drives are faster hard drives. If you have 10MB per track on a hard drive that rotates as fast as another drive that has 1MB per track, that's a 10x increase in throughput.

And, secondly, because making the mechanical parts of a hard disk move faster is much, much harder than sticking more data on the disk.

Personally I have 950GB of disk space on this PC and most of it is full; I could easily fill up another 1TB of disk space with all the video editing that I do.

Reply to MarkG

I agree with MarkG on this one.

There's a lot of recreational computer users that wouldn't need more than 20 gigs since all they do is surf the web and use spreadsheets.

There's also a lot of people like the tomshardware type that downloads huge amounts of data (legal and illegal, but i'm not going there). I know a guy who has over 1TB on optical storage and another guy who has 4.5 TB on optical storage as well.

These are the kinds of guys that need large drives.

I do disagree with the increase in drive size. If raid 5 was incorporated ( i forget who mentioned it, but YOU deserve the credit), then no worries.

I'd stick with a RAID card and buy 400 gig drives x12 and had 4.5 TB without worrying that one dead drive will mean i lose all my data.

Reply to Synthetickiller
- 0 +

Yes,
Normal people will still only need 20 gig for office apps.
I think that is what flash drives will be for. Apple is also switching to flash drives for the iPods - not many people have enough mp3s to fill a 60 gig 1.8" drive.
So the rest of us will get 400gigx6 RAID 5 and 1TB drives.

Reply to enewmen
Previous
1 2
Tom's Guide > Forum > Storage > Hard Disks > 1TB milestone reached
Go to:

There are 13 identified and unidentified users. To see the list of identified users, Click here.

Please mind

You are about to answer a thread that has been inactive for more than 6 months.
If you still wish to proceed, please ensure that your posting is original and does not duplicate or overlap any prior responses to this thread.

Add a reply Cancel
Google ads