Seagate Plans 37.5TB HDD Within Matter of Years

Ralph_19 writes "Wired visited Seagate's R&D labs and learned we can expect 3.5-inch 300-terabit hard drives within a matter of years. Currently Seagate is using perpendicular recording but in the next decade we can expect heat-assisted magnetic recording (HARM), which will boost storage densities to as much as 50 terabits per square inch. The technology allows a smaller number of grains to be used for each bit of data, taking advantage of high-stability magnetic compounds such as iron platinum." In the meantime, Hitachi is shipping a 1 TB HDD sometime this year. It is expected to retail for $399.

Comments

Terabits???

[Anonymous Coward]

It's bad enough that hard drive manufacturers are dead set on confusing people with 1,000,000,000-byte GBs. Do they really need to start throwing around figures in Terabits? Seriously, enough is enough...

Re:Terabits???

[FunkyELF]

Is that where the difference comes from? The title of the slashdot article says 37.5 TB then in the summary it says 300 terabit.

Re:Terabits???

[AikonMGB]

Through the magic of math: Tb / 8 = TB and so (300Tb)/8 = 37.5TB

/GASP

Aikon-

Re:Terabits???

[Yvan256]

The kilo/mega/tera/etc comes from metric, not the computing industry. A kilometer is 1000 meters, not 1024 meters.

I do agree on the "bit vs byte" part, though.

Re:Terabits???

[polar red]

bits&bytes is NOT COVERED BY the metric standard ! SO : 1 Kilobyte = 1024 bytes !

Re:Terabits???

[Anonymous Coward]

actually they are. according to IEC 60027-2
1 kilobyte (kB) = 1000 bytes
1 kibibyte (kiB) = 1024 bytes

come on, the original specs date back from 1999.

Re:Terabits???

[lewscroo]

Yeah, the HD industry really needs to stop doing this. I mean, with Terabit drives, you are going to be loosing huge percentages due to that stupid 1000 = 1024 logic they have. 1000 GB is going to end up really being closer to 930 GB

Re:Terabits???

[Alioth]

We should deprecate all the 1000-based metric suffixes dammit. Life would be so much easier if we all counted in hexdecimal - I was writing an asm routine for an 8 bit embedded machine to convert decimal to a short the other day, and if we only counted in hex, the routine would have been about quarter of the instructions that it came to!

Re:Terabits???

[qazsedcft]

It's the OS manufacturers that need to get up to speed. The definition of GB is 1,000,000,000 bytes. You're thinking of GiB. See http://en.wikipedia.org/wiki/Binary_prefix

Re:Terabits???

[egr]

Hate GiB MiB KiB TiB etc, they look ugly

Re:Terabits???

[Yvan256]

So, it would have been 4 lines of code instead of 16?

Re:Terabits???

[jZnat]

You mean 4 instead of 10 you insensitive clod!

Re:Terabits???

[AikonMGB]

8 b = 1 B, therefore 8 Tb = 1 TB (or 1 Tb = 1/8 Tb if you prefer)

Less confusing? You just stated that 1Tb = (1/8)Tb..

Aikon-

Re:Terabits???

[Fred_A]

How much is that in TiB anyway ?

Re:Terabits???

[kkwst2]

It's not much worse that what you stated, which could be interpreted as a byte is one eighth of a bit.

I understood what you were trying to say, but it wasn't clear. Try:

300 Tb * (1 TB /8 Tb) = 37.5 TB.

or to generalize:
x Tb * (1TB / 8 Tb) = x/8 TB.

Simple mistake (or abmiguous notation, if you prefer), but kind of funny since you were ribbing the parent about simple math!

Re:Terabits???

[blane.bramble]

Except, of course, that 1Kb = 1024 bytes was in use long before 1999...

Re:Terabits???

[Yvan256]

LOL, nice reply.

Re:Terabits???

[Zaatxe]

Here at work we say that the way to tell apart a programmer from an engineer is to ask him/her how much is 1K. If they say 1024, they are programmers, if they say 1000, they are engineers.

Re:Terabits???

[TheRaven64]

1KB was used to describe 1024 Bytes earlier than 1980. Standards that come along and re-define terms in common usage decades after their first use should be ignored, and are by everyone except those marketing hard drives.

Re:Terabits???

[COMON$]

Got a good chuckle out of that....thanks :)

Re:Terabits???

[swillden]

How much is that in TiB anyway ?

300Tb = 300 * 10^12 bits
= 300 * 10^12 / 8 bytes
= 300 * 10^12 / 8 / 2^40 TiB
= 34.1 TiB = 34,924.6 GiB

Any way you count them, that's a lot of bits.

Re:Terabits???

[noidentity]

When you're talking about data densities, you're talking about raw bits, not the nice error-corrected bytes the drive gives you back. It would be more confusing if they talked of raw bytes, when the actual byte capacity would be less due to error correction and formatting.

Re:Terabits???

[Jherek Carnelian]

1KB was used to describe 1024 Bytes earlier than 1980. Standards that come along and re-define terms in common usage decades after their first use should be ignored, and are by everyone except those marketing hard drives.

Or networks. What a rip-off that your gigabit ethernet card will only transmit at 1,000,000,000bps!!

Re:Terabits???

[Martin Blank]

Standards that come along and re-define terms in common usage decades after their first use should be ignored, and are by everyone except those marketing hard drives.

...and Slashdot pedants.

Re:Terabits???

[inode_buddha]

The metric system is base-10. Binary systems such as computers are in base-2. For the purpose of histronics both mensuration and computation are branches of Science, and the prefixes are equally valid. IMHO what people complain about is that marketing uses the "base-2" numbers to make things look bigger to a "base-10" mind.

Re:Terabits???

[f1055man]

Except GB = 1,000,000,000 bytes is pure nonsense. Giga means a billion? Sorry, this isn't a base 10 field. Football announcers don't say "hand off to Reggie Bush for a gain of 4 meters." Even Madden is not that retarded. Marketers and hard drive engineers need to stop applying a convention that means nothing in the computer science domain. I refuse to use MiB or GiB because they can't wrap their heads around the binary system. Actually they are both wrong. In decimal, giga is 10^9, mega = 10^6, kilo = 10^3 therefor in binary it should be: giga = 2^9, mega = 2^6, kilo = 2^3.

Re:Terabits???

[foamrotreturns]

It's not that any of them are dead set on it. It's a cold war. For example, if Seagate decided that it needed to get right with its consumers and started labeling its disks accurately, then Maxtor, WD, etc would have the competitive advantage because their drives of identical capacity would be labeled as having more space. Ultimately, it's the consumer's fault for not reading the fine print on the boxes he buys. If Seagate (or any other manufacturer) could trust its consumers to be informed enough not to buy into the 1000 byte Kilobyte farce, then they would be able to change. But the fact remains that the consumer is stupid and cannot be trusted to make that decision. Long ago, some marketing twerp at one of these companies figured out that kilo/mega/giga ... are all metric units measured in thousands increments and decided to use the metric measurement and put the "decryption" key (1 Gigabyte = 1000 Megabytes) in 2 point font on the bottom of the box. Unable to make a hard disk whose actual capacity matched this one's imaginary capacity for a similar price, all the other manufacturers followed suit.
So you see, it's not that any of them are trying to deceive. They're simply caught up in a game that some idiot started in attempt to make a fast buck. Now everyone has to suffer. Thanks a lot, marketing twerp.

Re:Terabits???

[dosquatch]

What do they say if they are software engineers? First they'd have to ask how much that is in ounces.

Re:Terabits???

[ottffssent]

Platter density is in terabits/sq in. Deal. That's the way it's specced. Research numbers are reported in bits because that's how the media works. When you add formatting, coding to limit long strings of duplicate bits, long Reed-Solomon ECC, etc., you lose a lot of those bits and get something that can more properly be expressed in bytes.

Do you whine about ethernet being described in bits? Ethernet doesn't deal with bytes - that's at a higher OSI level. Similarly, platters don't deal with bytes: that's the controller's job.

Re:Terabits???

[jo42]

I propose we move to an even better standard in storage space measurement: WLC (Whole Lotta Cr*p). So now instead of buying a 500GB HD, you buy a WLC x500 HD and so on.

Re:Terabits???

[bram]

Depends on who's asking.

Re:Terabits???

[quizzicus]

Could somebody explain to me why we can't just use 1 KB = 1024 bytes (and so on) for everything?

Re:Terabits???

[Novus]

Sorry, that last one should be 1 Tb = 1/8 TB. The scary part is that I proofread that to avoid making a fool of myself.

Re:Terabits???

[Wildclaw]

Yes, Giga is latin for billion (10^9).

So get off you high horse and admit that it is the computer scientists fault for trying to change the definitions of an already existing prefix system to fit their own domain.

Re:Terabits???

[kimvette]

Because of tradition and consistency.

No HDD manufacturer is going to want to advertise their 78.1GB HDD, competing against their competitors' 80GB HDDs, even though the 80GB HDDs do list the actual capacity in traditional, standard capacities in fine print. The current rating system is here to stay because convincing ALL of the hard drive manufacturers to make this change is next to impossible, and Hell will freeze over before ONE of them will adjust their rating systems without EVERY other manufacturer agreeing to immediately follow suit.

Re:Terabits???

[Lord Ender]

There are still people like you? Sometimes, the first idea is a bad idea. When a better one comes along, people switch.

If you have only ever worked entirely in the IT industry, you might not see the point. If you have done cross-disciplinary work, you will realize that redefining common binary prefixes in one field but not another leads to confusion and ambiguity where those fields combine.

Re:Terabits???

[Kjella]

Man, you really need to check out some more standards. Very many transmission standards use 1kB = 1000B. Video codecs and other streaming stuff too. In short, it's a mess that people get caught up in all the time, not just when buying HDDs. Besides by your very own argument, it never should have been 1kB = 1024B in the first place. The prefixes are univeral across all sciences and in daily speech (e.g. kilometer, kilogram), and it was computer scientists who "came along and re-defined terms in common usage decades after their first use". They created their own little sub-domain where it works differently than everywhere else. It's like trying to use a C++ class and discover that for this particular class, the + operator is doing multiplication instead. No matter how you twist and turn it, prefix overloading was a really crappy idea to begin with and it can either be fixed or remain broken, but don't kid yourself into thinking that it's anything but a very obscure exception to a common standard.

Re:Terabits???

[mdwh2]

There are still people like you? Sometimes, the first idea is a bad idea. When a better one comes along, people switch.

Yep, everyone has switched, it's just except him, and the OS used by most PCs on the planet.

you will realize that redefining common binary prefixes in one field but not another leads to confusion and ambiguity where those fields combine.

Not anywhere near as much as the confusion from redefining a common prefix in the same field...

Re:Terabits???

[dangitman]

Because it doesn't make any sense. "Kilo" means 1000, not 1024.

Re:Terabits???

[mdwh2]

That's not what the link you gave says, Wikipedia supports that GB can be used for 2^30 bytes. Yes, a new definition has been introduced, but it has not replaced the old one.

Re:Terabits???

[dangitman]

Which just reinforce the GP's point. The OS/software manufacturers are actually using GiB, MiB, etc. The problem is that they are mislabeling them as GB, MB, when they are not. If they would actually switch to using the terms accurately, there would be no problem. Is there any reason to use that 1024-based crap these days? You'd never have to hear those ugly terms again, if people used the 1000 based notation consistently.

Re:Terabits???

[mrchaotica]

How about they just make drives that actually are 80GB (in the common reckoning -- i.e., 80*(2^30) bytes instead of 80*(10^9) bytes)?

Re:Terabits???

[mrchaotica]

Nothing, because they're elitist idiots. As a civil engineer and computer scientist, I'll tell you this: there is no such thing as "software engineering!" If there were, the liability settlements alone would have killed off the entire industry years ago.

Re:Terabits???

[BillyBlaze]

How many Libraries of Congress is that?

Re:Terabits???

[Macka]

Using giga, mega and kilo in decimal makes sense because they are all (decimal) Greek units that existed several hundred years before electronic computers were invented. It's applying these terms to binary that was (is) a stupid idea in the first place. The scientists at the time should have exercised more imagination and created something unique and non contradictory. Thus the creation of Gibibytes (GiB), Mebibytes (MiB) and Kibibytes (KiB). They are the correct terms now and its only laziness that has stopped them from becoming ubiquitous. Schools, Colleges and Universities should make a point to only using the correct terms so that the next generation of scientists/programmers/admins get it right.

Re:Terabits???

[polar red]

the definition of "kilo" being 1000. Nope. Kilo is NOT defined as 1000. Kilo is only defined in conjuction with metres,liters,grams,.... in the SI-system as a 1000 units. In conjuction with Byte, it has been defined as 1024 units, since at least the sixties.

Re:Terabits???

[dangitman]

Nope. Kilo is NOT defined as 1000. Kilo is only defined in conjuction with metres,liters,grams,.... in the SI-system as a 1000 units.

Which is what I said. It should have been pretty obvious I was talking about 1000 units as a modifier.

In conjuction with Byte, it has been defined as 1024 units, since at least the sixties.

By who? And why? It makes no sense, and has never been an established standard. Just a stupid quirk that some idiots decided to use to screw with the rest of the scientific world.

Re:Terabits???

[polar red]

and has never been an established standard. By the whole industry(except HD-makers) and academic world as a de-facto standard.

Re:Terabits???

[dangitman]

In other words, not a real standard. If the whole building industry redefined a metre to match the most common length of lumber they use, then they would be wrong too, and should reform their practice. Just because a bunch of people do it, does not make it right. I think we should strive to improve ourselves, not get stuck in old habits.

Re:Terabits???

[dangitman]

No shit! I already kn ew that. It is still not a valid reason for redefining "kilo" as 1024. Kilo means 1000, not "around about 1000." Aren't scientists and engineers supposed to be precise?

Backup Solution?

[LibertineR]

I want to see the tape drive for that thing, Bitches.

Re:Backup Solution?

[Beryllium Sphere(tm)]

Easy. Delegate the backups to your worst enemy.

At $399, you could buy a bunch of them and use them in a rotating backup, periodically sending one offsite. Or use it as the destination for nearline backups of everything else on your network.

Re:Backup Solution?

[LibertineR]

I swear to God this is true. I had a client ask me to create two partitions on a 500G drive, which was loaded with 200G of medical insurance claims. When I asked why, he said that although he didnt want to buy another drive, he understood the importance of having a backup for his data.

I sprained a rib, choking back a laugh.

Re:Backup Solution?

[lonechicken]

I wouldn't want to wait for the defrag time or the spyware/antivirus scanning times on these drives.

Re:Backup Solution?

[Anarke_Incarnate]

Backup tape density grows as well. LTO 2, which my company uses can store 200N/400C GB per cart. The truth is, our source code is so easily compressed that we get 500N/1000+C GB per cart when dealing with it. LTO4 should be out soon. I can only assume that in 5yrs tapes will be easily dealing in 20TB uncompressed.

Re:Backup Solution?

[Anonymous Coward]

um, you could use linux/macosx/bsd/solaris

Re:Backup Solution?

[cdrudge]

It depends on what the "backup" is for. If it's for disaster recovery, then you are right. But if it was a on-line backup in case of an "Oh shit I didn't mean to delete that" type of thing, then dual partitions can make sense.

Re:Backup Solution?

[Dr_Barnowl]

That's not entirely spurious.

It is at least better in some ways than those people who think that a RAID array is an excuse not to have a backup.

Hardware failure is not the sole cause of data loss - in fact, I'd be prepared to believe that data loss down to "software" failure is more common, be it a human being who accidentally bangs "delete", or a virus, or bad software corrupting the records, etc..

I'm not saying it IS more common, but it's certainly plausible.

However, the attitude here displayed by someone who is in insurance and thus supposed to understand the benefits of a good fallback plan..... I would recommend he goes for a RAID of some sort, with a hard drive caddy that receives the backups with the drive cartridges rotated offsite periodically.

Re:Backup Solution?

[LibertineR]

No, the dude was planning to schedule backups from the Microsoft applet to the other partition every night, thinking that if the first partition failed.......

I kid you not.

Re:Backup Solution?

[Anonymous Coward]

Separate partitions can isolate logic errors though, which are the most common cause of data corruption.

Re:Backup Solution?

[brainspank]

That's pretty funny to imagine, but there's really no need for tape except for very long-term storage. It's very economical nowadays to keep a large (enough) RAID array of disk for an short- and mid-term backups, in combination with WORM/DVD/tape for long-term backups. Anything that doesn't need to exist longer than 6 months for us, doesn't ever go to tape. You get great restore times, too. It's the "Emeril" restore - BAM!

bs

Re:Backup Solution?

[segfaultcoredump]

Ok, so you are spending $399 for a 1TB drive. Compare this to a 400G (uncompressed) LTO-3 tape @ $50 per tape (price is good as of yesterday when I ordered another 100 LTO-3 tapes). Your drive is about 3x as expensive.

The tape is still cheaper. It also takes up less space on my shelf and I can drop it and not worry about loosing anything.

I am looking at these drives for the front end disk array that I use in my d2d2t setup (disk -> disk -> tape). Given about 40 of them I can keep 2-3 weeks of backups online in the disk and then destage to tape for the offsite vault and archive backups. This way restores of recent data is almost instant (no need to mount and seek to the spot in the tape), but the old archives cost me less and I save on power and cooling (the tape library expansion modules take no additional power. its just a shelf with tape slots).

Its not an either/or choice. Most folks with any real amount of data to backup use both.

Re:Backup Solution?

[clydemaxwell]

lord knows you never fsck or clamscan in linux..

Re:Backup Solution?

[dosguru]

They cost about $30,000/ea. Our solution is the 500GB/tape raw, STK T10000A drive, they write at 120MB/s raw on our 2Gb FC and get about 1.67:1 compression on our 400 servers. The T10000B will use 1TB raw tapes and have 4Gb fiber interfaces.

Re:Backup Solution?

[TheRaven64]

How much is the tape drive though? If you are only backing up one or two hard drives, and only need one backup then it still might be cheaper to use other hard disks.

Re:Backup Solution?

[Kjella]

Separate partitions can isolate logic errors though, which are the most common cause of data corruption.

Oddly enough I never seem to run into that, and yes I do have parity checks for quite a few files. Modern drives seem very good at realizing a sector is about to go bad and remap it. What usually happens is that one day you try to boot up and your hard disk is a brick. Either it's the electronics or the motor or the bearing or whatever took a nose dive and the disk is dead. No, RAID1 is not a proper backup for oh so many reasons but it's automatic - the problem with almost any other backup solution is that you need to actively start the backup. At least I get annoyed if my disks go nutty or my bandwidth gets eaten or annoyed if I have to put in DVDs or tapes or attach that external HDD.

Re:Backup Solution?

[GeffDE]

While the guy probably didn't know this (hell, I didn't until it happened to me), two partitions can make or break a hard drive. On my laptop, I had one partition for Mac OS X and one for linux. One day, OS X decided it didn't want to boot. However, Linux booted fine. It turns out that the hard drive was failing and the first place it failed (that I noticed...) was right over OS X's boot partition. So, if the failure of the harddrive just starts and it fscks just one of the partitions, the other will still be readable...

Re:Backup Solution?

[Ironsides]

Ok, so you are spending $399 for a 1TB drive. Compare this to a 400G (uncompressed) LTO-3 tape @ $50 per tape (price is good as of yesterday when I ordered another 100 LTO-3 tapes). Your drive is about 3x as expensive.

You are failing to include the cost of the LTO-3 Tape DRIVE. Which, in itself, will cost around $3000 for a cheap one (thanks, CDW) and require a SCSI connection.

So, while the LTO-3 Tapes may cost 1/3rd as much, they still require the several thousand dollar LTO-3 drive and a SCSI connection which most computers do not have. So, for lower amounts of data, the drives are cheaper.

Re:Backup Solution?

[bladesjester]

In that case, should there be a catastrophy (fire that takes out the building, etc), you'd be dead in the water since your most recent backups available would be 6 months old.

Part of the game of backups is planning for the worst but hoping it doesn't happen. Make regular backups, send a copy offsite, and go on with life. That way, you aren't left going "oh crap" when your machine is torched.

Re:Backup Solution?

[dilvish_the_damned]

Thats better than when I was told to move a clients windows 3.1 swap file to ram for a performance increase. Its an old joke now, but I think that Kim Komando might have suggested it about the same time she was raving about the merits memmaker.

Re:Backup Solution?

[asuffield]

You are failing to include the cost of the LTO-3 Tape DRIVE. Which, in itself, will cost around $3000 for a cheap one (thanks, CDW) and require a SCSI connection.

And a maintainance contract, because the damn things are fragile piles of barely contained self-destruction, and break at least once a year. And an intern to care for and feed the stupid thing with tapes, because you can't afford to buy a robot to do it, the sysadmins threatened to resign if you made them do it, and the users cannot understand instructions like "If you have deleted all your files, tell us BEFORE two weeks have elapsed so that we don't have to go and fetch a tape". Estimated annual cost, around $10k (assuming that you can get some work out of the intern the rest of the time, so only part of their salary is counted).

Tapes aren't something you want to buy and run yourself - they're something that you want to get somebody else to deal with for you. At this point, off-site data vault companies start to look better. (Send them your data via VPN or just courier a drive to them every week)

Re:Backup Solution?

[segfaultcoredump]

Ok, here goes:

In my environment, I have 500+ LTO-3 tapes (and I'm purchasing about 100 more each month). That cost about $25,000. I also have a Library with two LT0-3 drives. $3000 is for a very cheap one. Lets use a real library and get an ADIC i500 w/ 2 FC LTO-3 Drives. My cost for that library with a 4Gb dual port FC card was $30,000. Total cost for this setup is $55,000. Note that I dont need to keep all of the tapes in the library (most of them sit on a shelf since it is cheaper than the expansion slots for the library).

In order to be able to archive 200TB of data, I would need 200 of these drives (assuming no redundancy). That comes out to about $80,000 for the drives. The price does not include the price of the enclosure to actually connect the drives to the backup server. For the sake of argument, lets assume that cost $0 just to make the drive kids happy. Lets also assume that you dont pay anything for power to keep the drives spinning. Lets also assume that if you drop a drive from 6' nothing will happen to it.

Notice how the tape solution is $25,000 cheaper. The break even point is around 275 LTO-3 tapes worth of data, after which the tape solution becomes cheaper (using my overly simplified calculations that favor the drives).

Please note that the best solution depends on your environment. I run through about 20TB of data per week for backups (I consider that a small to miszied environment). We need to send tapes offsite (weekly), and my datacenter is limited on the power and colling standpoint. If you only have 500gb of weekly backups, then go ahead and toss 15 of these into a disk tray and have at it.

Notice how I still use disk as a front end. Its the only way I can feed my two LTO-3 drives. There is no way I can pull 200MB/s of data over the network from the clients and keep the two drives streaming. It also reduces the number of tape drives I need in order to run 200 backup jobs each day since I can multiplex 20 jobs at once to each disk array to try and drive overall system throughput up. But it is not cost effective for me to leave the data there.

And for the folks who think that shipping 20TB a week over the network to an off site company is a good idea... thats 34MB/s at 100% utilization. The cost of that link alone would pay for the entire tape system after a year. Never underestimate the bandwidth of a truck loaded up with tapes. :-)

Re:Backup Solution?

[brainspank]

you're right. however, nobody ever said that the disk/tape/dvd/worm/blu-ray needs to be in the same location as the original data... If my life/business depended on it, I would certainly keep it elsewhere. In other words, the media isn't important, it's the policy that keeps you "safe."

Re:Backup Solution?

[LunaticTippy]

While 2 partitions on 1 failing drive might save your hiney on rare occasions, I think it's safe to say only a retard would think of this as a backup.

A backup would be a 2nd physical drive. Anyway, this one time I recovered all the essential files off a dying drive with disk commander or whatever. That doesn't mean backups aren't important.

Re:Backup Solution?

[LunaticTippy]

Another advantage of tapes is cheap multiple backups. When the drive is the pricey part, you can afford to have 2 or 3 sets of tapes and cycle through them so you always have multiple backup sets.

Re:Backup Solution?

[Antique Geekmeister]

Since the biggest risk to such data is user error, not hardware error, it's not unreasonable. Slap in a hard-linked backup system such as "rsnapshot", and you can considerably ease recovery tasks.

But he definitely needs another drive on another system for that, given the amount of the data and the risk of catastrophic failures.

Re:Backup Solution?

[totally bogus dude]

He said "defrag", which is not in any way related to "fsck".

And I never do a full-system AV scan, but then again I don't run random binaries from the internet on my computer, either.

That's great.

[Aladrin]

That's a great amount of storage and a great price, but what about some REAL information: Speed, heat, power consumption. If for the same price I can run 4 250gb drives and save on heat and increase speed, this doesn't make sense to do. If I can run 6 and RAID them, and gain security, it really doesn't make sense.

The largest drive in the world isn't any use to me if it's slower than a 3.5" floppy or I can use it to replace my space heater.

Re:That's great.

[ImdatS]

Just quickly, the specs I found for the Hitachi Drive:

- 5 discs, two heads each, rotating at 7200 RPM
- 1070Mbps transfer rate
- 8,7ms avg seek time
- 4,17ms avg latency
- around 9 watts power consumption while in "inactive-mode" (NOT reading or writing)

Hope this helps

Re:That's great.

[mseidl]

if I did my math...

300tb = 37.5 TB?

I hate how people switch up the bits and bytes. It's very misleading. Remember /. kiddies that 1 bit = 8 bytes.

Since this being /. , you wont RTFA or RTFWP(Wikipedia)

Re:That's great.

[Anonymous Coward]

Umm...
At the risk of sounding sarcastic (I know that never happens around here)

I'm pretty sure there are 8 bits in an octect (which for all intensive purposes is a Byte)

Re:That's great.

[phasm42]

Remember /. kiddies that 1 bit = 8 bytes.

Read that again carefully.

Re:That's great.

[welcher]

OK, so it's not "intensive purposes", it's "intents and purposes". Makes a bit more sense, no?

Re:That's great.

[mseidl]

oh Shit. Way to go me!

1 bit = 1/8 of a byte. Thats right!

Re:That's great.

[Achoi77]

That's a great amount of storage and a great price, but what about some REAL information: Speed, heat, power consumption.

I don't think marketers will even care about all that, their only benefit will be the low cost, high density factor. Why?

Because SSD are coming. And HDD manufacturers know this. SSD are such a humongous threat to their business, HDD are going to be obsolete tech the moment SSDs come out, so that is why manufacturers are playing their last trump card: the fact that they can hold gobs and gobs of data. Their only hope is to find a foothold in the archiving business.

that is untill we start seeing cheap high density SSD. But that won't be for a while.

Re:That's great.

[clydemaxwell]

You do mean 1 byte= 8 bits, yeh?

Re:That's great.

[tehcyder]

Hey, you had a 50/50 chance of getting it right, you were just unlucky.

Reliability

[nten]

They mention using HAMR to increase stability. Does anyone know if it could be used without bit patterning to increase the reliability of current large drives? You know, the ones with 2yr life expectancies or less.

Heated platters?

[140Mandak262Jamuna]

Well, I like my pasta primevera on heated plates but I am not so sure I would put 37 TB of my data on platters that get heated repeatedly, till some independant testing shows the durability of the data.

Re:Heated platters?

[oohshiny]

HARM uses highly localized and short-term heating. I doubt the platters as a whole are noticeably affected.

Re:Heated platters?

[Huggs]

HARM? How is HARM Heat Assisted Magnetic Recording? Shouldn't it be HAMR?

Re:Heated platters?

[CityZen]

Do you want to HARM your data or do you want to HAMR it?

Frankly, neither one sounds very appealing to me.

Re:HARM

[aardvarkjoe]

Although amusing, HARM is not an acronym for "Heat-Assisted Magnetic Recording." Looks like Zonk didn't even read the summary again, much less the article...

Re:HARM

[abrinton]

Actually, it's HAMR. Which isn't much better.

HAMR those terabits. All 8 of 'em.

Re:HARM

[Joebert]

If they could prepend somthing that started with C, I think it would work like a CHARM.

Re:HARM = Anti Radar

[flyingfsck]

High Speed Anti-Radar Missile - Run for the hills!

Unit of measure

[sphealey]

If this is to be a tera_BIT_ drive then I believe the headline should read "Tb" rather than "TB".

sPh

Re:Unit of measure

[Cctoide]

37.5TB = 300Tb. TB is a "rounder" unit and as such is more suitable to a headline, although it's still a bit confusing.

Re:Unit of measure

[mosschops]

If this is to be a tera_BIT_ drive then I believe the headline should read "Tb" rather than "TB".

And ironically, there's only 1 bit difference between those 2 statements :-)

any relation to Cringely's disk?

[Globulatrix]

I wonder if this has any relation to the Robert X. Cringely hard drive.

Bob's Disk Drive:
http://www.pbs.org/cringely/pulpit/2006/pulpit_200 61026_001143.html

The Wired article mentions Iron. Cringely's mentions Stainless Steel?

Re:Reliability?

[flyingfsck]

Well, if it doesn't work, just use a bigger HAMR...

OS/BIOS

[s31523]

I hope OS and BIOS manufacturers are listening... I'd hate to drop 400+ on a hard drive to have it seen as 1/3 of the actual size by either BIOS or the OS.

Re:OS/BIOS

[jimicus]

No sane OS has paid the remotest bit of attention to what the BIOS thinks the disk is in years.

Re:OS/BIOS

[drinkypoo]

No sane OS has paid the remotest bit of attention to what the BIOS thinks the disk is in years.

That's very true; this is why I had to use a disk extender with Windows XP on a computer I set up for someone - it wouldn't properly detect the size of the disk.

Re:OS/BIOS

[Kjella]

Well, if you're only a fairly recent machine that passes the 137GB/128GiB barrier fine, then I doubt you'll have much problem. ATA-6 defines 48-bit addressing and has a theoretical maximum of 144PB/128PiB. Modern file systems can also access something like 2^63 bytes = 512PiB. Looks like clear skies for 37.5TB HDDs from what I can tell.

Re:OS/BIOS

[jimicus]

If it's LBA48 support you were missing, Microsoft did release a patch for XP to support that.

Re:OS/BIOS

[drinkypoo]

Unfortunately the windows installer was misdetecting the disk... so unless the patch is somehow applied to the installation media before installation, and not only fixes the installed windows but also the windows installer's copy of NT that you boot from the CD, I fail to see how it would help...

Re:OS/BIOS

[jimicus]

I imagine you'd need to use a copy of Windows which had the patch pre-installed - either slipstreamed on (does slipstreaming patches like that cause them to work at install time?) or a post-SP2 copy of the XP install CD.

Re:OS/BIOS

[drinkypoo]

AFAIK the patches only apply the installed OS, as they are merged into the install files, and do not affect the version of NT which is booted from the CD and from which the installer is run. I assume you would need the post-SP2 version of the CD, which would hopefully have the patch rolled in. I've never seen one of those CDs though :)

Re:OS/BIOS

[Antique Geekmeister]

You still need a 64-bit controller. 32 bit controller hardware still only gets you up to roughly 31 bits of disk size, or 2 Terabytes. Don't expect to see 64-bit controllers for desktop operating systems anytime soon, although there is certainly a demand for them in larger RAID arrays.

Re:HARM

[Lord_Slepnir]

If you've ever had to use Seagate hard drives, you'll know that HARM is an extremely appropriate name for them.

We had a batch come in for some IBM e-servers, and a third of them died within 6 months. Absolutely disgraceful. The ones we have running Hitachi hard drives are all still going.

Re:37.5TB HDD

[Zenaku]

How does the capacity of a drive have anything to do with its seek time? Seek time is a function of how quickly the read arm can cross the radius of the platter, and to a smaller degree how fast the platter spins. The article claims they will be increasing storage density using this HARM thing so that more bits can be stored on the same amount of surface area. Seek time should not change significantly unless they make the platters larger, or spin the drive at lower RPM.

product looking for a market

[cliffski]

Ok, so on the more general point of high capacity 3.5 inch drives, Does anyone really need these? In my experience, PC hard disks are already way too big. A friend of mine uses his 100 gig drive for some emailing, websurfing, playing a few games, and music playback. Last time I checked his PC it was over 85% empty. And most of the space that was consumed was the O/S.
All a bigger drive gives joe average is a longer defrag time, and longer search time. I'd hazard a guess that 80% of current domestic end-user drive space is currently empty.
Sure, many slashdotters will have filled their disks with all manner of stuff. I'm a developer, and the obj files alone for games stretching back 10 years certainly take a up a huge chunk of my disk, but we aren't average joes.
I'll get a new PC next year for vista (I need it for checking games compatibility) and no doubt it will come with a 500-1000GB drive as standard. I'd rather it didn't, I've got by for years with my 80gig friend here. If theyt *really* want to innovate on disks innovate here:

Power consumption (esp with electricity prices going menatl as they ahve in the UK)
Seek Time
Cost

Why innovate on capacity? it's the one major metric that most people have stopped caring about. I'm not being a luddite, for a long time disk capacity *was* a major issue, and we regularly ran out of space. I think that time is over.

Re:product looking for a market

[lonechicken]

I've got two 300GB hard drives on one of my computers. There's "only" 85 Gigs left on one drive and 5 Gigs remaining on the other. And I regularly clean out games I don't play anymore, and have a separate computer for testing out MSDN stuff. So, yeah we're always going to need more.

Re:product looking for a market

[Prof.Phreak]

Data centers spend millions (literally) on storage. Try pricing a few hundred terabyte solutions, and you'll see.

Besides, if you could store all of music/movies/images that where -ever- created on your home drive (not just those copies of libraries of congress), why not? I'd certainly wouldn't mind having all that storage---cheaply.

Re:product looking for a market

[William_Lee]

Two words... p0rn and piracy...

Re:product looking for a market

[tom17]

Does anyone really need these?
  This and a quite from the article -

Although 1TB of storage on a single drive will be alluring to some users, IDC's Rydning sees only very specific demand for that much storage. "For consumers, we still think the big hard drives are mainly for niche applications," says Rydning. "There's going to be a certain minority of PC users and video recorder enthusiasts who will want to have the highest capacity available. And in those markets, a high-capacity drive is valued. However, the vast majority of PC users are still serviced by a one-platter, 160GB hard drive."

Consumers' increasing accumulation of digital personal data is, not surprisingly, driving the need for high-capacity storage. "As people amass their own personal memories, either in photographs or video, hard disk drive storage is one of the best, lowest cost ways to store and retrieve that type of data," says Rydning. - both just remind me of the same shit that people harp on about every time new breakthroughs are made in hard drive (or any other, I guess) technology. I remember them saying the same thing about 80GB hard drives.

People, please! The computing world advances, people use more hard drive space, people DO end up using it.

Re:product looking for a market

[z0ot]

Enterprise environments desperately need better storage density. For example, increasing government requirements for long term record storage mean that IT departments in the medial industry are increasingly stressed. As another example, the explosion of online media distribution (YouTubes, Napsters, and the like) require insane amounts of storage. You can't just keep making the data centers bigger. At some point you must increase the amount of storage each location can hold.

The home user is just one customer for hard drives, and probably not the one targeted for this capacity of hard drive (yet). However, think about the amount of storage required to store large amounts full-quality HD video on a DVR and you might see that in a few years time even the casual home user might require this amount of capacity.

Not so long ago a 10MB hard drive was considered huge and hard to fill to capacity.

Re:product looking for a market

[Zenaku]

It's still a major issue for me. You're right, I'm not an average joe when it comes to storage needs, but does that mean that nobody should produce a product that fills my need? My 1.2 Terabyte RAID array is full, and I am currently wondering how the hell to add more storage and migrate the data without simply building a whole new machine.

The innovation in capacity and density is driven by the needs of enterprise users, and atypical users like me. The advances that come of it are then incorporated into lower-end drives as well. The reason that you start to see 100GB drives being the lowest capacity you can find is not because nobody could get by on less, it is because it would cost more to keep producing drives using the older technology -- each leap forward in drive technology has to be accompanied by retooling of manufacturing equipment and process, and it doesn't make a lot of fiscal sense to keep producing lower capacity drives if they cost as much or more to make as a newer one with higher capacity.

Re:product looking for a market

[bockelboy]

Does anyone really need these?

At work, I'm building a 200 TB storage system for a particle accelerator. It is a small part of a much larger grid that will eventually need about 3 - 5 petabytes of online disk space, and about twice that much on tape for backups.

I see myself buying a lot of these when the enterprise version comes out. Heck, with current systems taken into account, I'd only need about 3 SATAbeasts (48-disk enclosure) with these suckers in it.

Rule of thumb in computing: There is *always* someone out there needing more capacity, whether it is CPU, memory, disk, or bandwidth.

Re:product looking for a market

[ivan256]

Datacenters don't necessarily want larger disks. Frequently, they are performance oriented and are more interested in spreading their dataset across a larger number of spindles for increased performance. They end up using terabytes of capacity for gigabytes of data. Seagate in particular has shifted their roadmap from capacity to performance in their enterprise and business class products. High capacity is reserved mostly for end users.

Re:product looking for a market

[NSIM]

Depends on what you need to store, I've already got around 2TB of storage scattered around my house, and I seem to be like most corporates in that it's doubling every year. I have no doubt that I'll be looking at 1TB drives real soon and that larger drives are going to become increasingly attractive for me. I daresay much the same arguments were made about 10MB, 100MB drives, 1GB drives, 100GB and so on yet markets for these emerged pretty quickly.

Re:product looking for a market

[bazorg]

it's the one major metric that most people have stopped caring about

perhaps those 5% of people with odd requirements outspend the other 95% of people who don't need bigger disks?

Re:product looking for a market

[Urban Garlic]

Consumers will want significant capacity once media-center PCs with downloaded video become common, and this day is coming very soon. People want a la carte TV, and Netflix and iTunes and amazon and whoever else are gearing up to deliver it. Your children's equivalent of your DVD collection is going on those disks.

Re:product looking for a market

[Bob of Dole]

I could use two.
So yes, they're needed. Just because you might not need one doesn't mean there isn't any one who needs them.
(My multimedia friends would just about sell their souls to get one of those. Imagine all that high quality video you could store on them!)

Re:product looking for a market

[Alioth]

The same technology (which fundamentally is increasing the density of the data on the platters) can be used for making physically smaller, less power consuming drives. For example, a Microdrive sized disk with the capacity of current drives in desktop systems. This could go into a device like a HD camcorder, and have space for plenty of full quality uncompressed video.

Re:product looking for a market

[maxume]

The concept of having all of your music instantly available and all in one place doesn't quite sink in right until you actually do it. And it spreads like a virus. I had it in ~1999, and now, all my brothers have it, and some of their friends (some because of them, some independently). With media extenders hitting reasonable prices(especially audio) and wifi becoming really common, more and more people are going to be needing lots of storage. I guess people that don't care much for music won't, but I suspect that still leaves a pretty big market.

On the other hand, the 80 gigabyte disk in my mom's laptop is probably big enough that it will die before she fills it up with pictures. I've called it an 'infinite' drive as far as her purposes go.

Re:product looking for a market

[EvilNTUser]

You're a luddite because your complaint is senseless:
1. As long as someone needs the space and is willing to pay for it, increasing space is good regardless of whether the majority needs it.
2. Companies are working very hard on your needs as well, so you aren't being neglected.

In addition, you're also ignorant, because consumers are already buying HDTV video cameras. As long as people keep breeding, there will be massive amounts of new footage - and yes, for completely normal people.

Re:product looking for a market

[Dr_Barnowl]

I presently have 2 300GB drives in my media server, 500GB of which is given over to DVB-T (television) streams and compressed videos. They cost me about $150 (£75) apiece.

Capacity is important to me, therefore, as that partition is perpetually about 98% full (although I suspect that it also be full if it was three times the size, I'd just keep more of the good stuff for longer).

But yes, price is also important. The drives I bought were in the "sweet spot" where the cost per GB was low. The higher densities mean that you can make drives with fewer platens and importantly, heads. Heads being the most expensive part of the drive, and platens being the part that makes a drive consume more power, cheaper, quieter, less hungry drives should be a natural side product of gargantuan behemoths that merely consume the same amount of power as current models whilst costing around the same but holding an awful lot more data.

The huge data sizes are just a consequence of the 3.5" form factor. Conversely, 2.5" drives are now getting respectably capacious - today I bought a 160GB bus powered drive that rests comfortably on my open hand. Enough to back up my laptop, tote my entire music collection around on, and store a fairly heft wodge of whatever else I fancy.

Incidentally, larger drives speed up data operations, not the inverse. You are confusing a bigger drive with a bigger filesystem - a larger filesystem will take longer to defrag, true, but putting it on a larger physical drive will almost always increase performance as

  * The data rate past the head increases
  * The latency decreases (the increased density means the head travels a shorter distance to go places
  * Fragmentation will decrease as larger drives have larger expanses of contiguous free space.

Plus, for your example of search taking longer, not only does putting it on a larger drive improve seek times and data rates, but gives you enough space to put on index on as well (and the indexing will run faster because the drive runs faster, etc, etc). Yes, if you start putting more data in your filesystem, things will take longer, but that's only made POSSIBLE by a larger drive, it's the users choice, not the drives fault.

Re:product looking for a market

[rgmoore]

Ok, so on the more general point of high capacity 3.5 inch drives, Does anyone really need these?

Everyone? No. Anyone? Yes. Joe Consumer, who mostly uses finished files that somebody else has generated for him, probably won't need a hard drive that big. Frank Producer, who makes his own stuff that Joe will eventually use, wants all the space you can provide and then some. Applications like digital photography and HD video can generate enormous quantities of data, even after using efficient compression schemes. Processing that data can chew up even more space than the original data files. Serious users always want more space, and if you give it to them they'll figure out new ways of using it.

It's also important to consider that the higher densities needed for those ultra-capacity disks can also help out other users. Maybe a 37.5 TB 3.5" disk is overkill for your application, but that same technology could be used to make a 10+ TB 2.5" drive or a 1+ TB 1" drive. A high capacity, ultra-small drive could be just the thing for the next generation of high portability laptops and tiny form factor desktops.

Re:product looking for a market

[jedidiah]

Nah. Monster drives just make up for the fact that DVD jukebox technology is overpriced and primitive.

Plus, once data is divorced from physical media it becomes much easier to do interesting things with it.

Re:product looking for a market

[bill_kress]

Actually a larger hard disk can only help with defrag times and won't effect search times (assuming the same amount of data)

But your point on Why Innovate on capacity is a good one--moreover why innovate only on capacity.

How about a hard-drive with built in RAID? Spread writes across all platters so that if one glitches you get the others. Still, this is not optimal because you can't replace a platter--but still, can we work on using that extra space to improve reliability?

Re:product looking for a market

[Rolgar]

I have a DVR with (I think) a 160 GB drive, at 4.6 quality (480p?). It holds 72 hours of content. Now, to have high def quality, even with better compression, I'd probably only have 20 or 30 hours of space. I could fill that in a month.

In 5 years, when I'm ready to go high def, I'll have to replace my DVR, and I'd like to be able to rip movies for my kids to the drive so they don't have to EVER handle the disks. If I have 30 times the space, I should be able to store every Disney/Pixar/other cartoons and TV shows we decide they can watch on the machine, and they'll never need to handle the disks. Hit the library button, scroll to the program they want, hit enter or play, and they'll be watching their show in 30 seconds from when then machine turns on (this is what I have on my current machine), much faster than finding the disk from the DVD shelf, getting the box open, getting the machine open, waiting for the disk to load, wait for the FBI warning, previews, and finally the menu, and 5 minutes later, getting to the start of the program.

As for my desktop computer, I've been storing collecting photos for just over a year now with one child, and we've got about 1.5 GB of content. I can imagine that collection approaching 100GB easy, and maybe more. Video could easily by 100s of times larger, although we haven't done any yet, and haven't really decided if we want our kids living in the past watching videos of themselves all the time, but when it comes to multimedia content, I can definitely see where someone would be capable of filling these drives over time.

Besides, there is little profit in the hard drive market, and they can charge more for the higher capacity for little additional cost. The price difference between a 250GB drive and a 500GB drive is more profitable than the initial cost of the 250GB drive, because they've already taken into account all of the physical components such as the casing, drive electronics, motor, head, and half the platters. Also, the hard drive manufacturers can only differentiate themselves on 4 factors: price, reliability, performance, and size. Most consumers don't care about or understand performance or reliability in computer parts, and that leaves size and price. That's what the companies have to advertise on, and since the prices can only go so low, sizes have to go up. If one company chooses not to compete, soon the others will be selling a drive 2-4 times the size at the same price, and they'll have no customers. This is the competitive hard drive industry, and we'll continue to see better products at lower prices, and we'll continue to like it. OEM shipped PCs will have TB disks if this is true.

A few months back, Cringley wrote (http://www.pbs.org/cringely/pulpit/2006/pulpit_20 061026_001143.html) that he knows some people in the industry, and that platters were about to be replaced by much stronger materials, that would bring about lighter and thinner platters, which would allow 3 times as many platters, lower powered motors to drive the platters, less energy, and these drives would have higher reliability than current models, and the cost would go down significantly. I wonder if this article is referring to the same transition that he was getting at.

Anyway, I look forward to a future where I'll have a file server the size of a deck of cards, and holds every movie and picture my family owns. It will be a great time to be a computer geek.

Re:product looking for a market

[tehcyder]

I'd hazard a guess that 80% of current domestic end-user drive space is currently empty.

Probably true, but this won't be a problem for geeks, as the amount of pr0n is practically infinite.

Re:product looking for a market

[Jeff DeMaagd]

Your goals are mostly contradictory, to resolve it, you either have to make choices or wait several years. Seek time increases power consumption. Lower power consumption costs more and reduces seek time.

A hard drive isn't getting any simpler to make, but they aren't really getting any more complex. There's only so low you can get on the labor and manufacturing costs. As you can get a useful hard drive for $50 now, why complain about the price? Reducing capacity doesn't really save much.

Desktop hard drive power consumption might average 10 watts. A notebook hard drive takes about two watts. A notebook hard drive is slower, smaller and more expensive though.

Re:product looking for a market

[ucblockhead]

Used to be some people liked big disks so they had space to rip all their CDs...we've passed that stage, but are rapidly approaching the stage where some people will like big disks so they have space to rip all their DVDs.

My home server currently has 900 gb worth of disks and is maybe 70% full.

Re:product looking for a market

[mihalis]

Ok, so on the more general point of high capacity 3.5 inch drives, Does anyone really need these?

Sure, anyone doing High-def digital video. Suitable high-def camcorders start under $1k, filesize is approximately 35GB/hour, before you start editing etc. Similar problem if you use your computer as a PVR. I use a Elgato EyeTV to record formula 1 motor racing, and I also shoot my own high-def digital video. The 400GB hard disk on my powermac is almost always fully and I'm starting to fill up a second drive now.

some of my blog entries on this

Re:product looking for a market

[kinglink]

Yeah because you're who they want to sell it for. Oh wait no you're not.

For one thing I have half a terabyte on my computer, and it's always filled up. I work at a company who works on 360 games (full scale ones, we have one out now that's doing reasonably well) and our projects are easily 20 gigs + Builds. So we might not need a full terabyte each. However our server could use it, anything that controls data or source, has daily backups can easily use it. Financial systems record keeping easily eats up space as does a number of other major business. Just because YOU the little programmer can't fathom a use, there's easily uses for more space. You act like people have stopped caring about it, but you talk about 100 gigs like it's nothing. A couple years ago 100 gigs was a lot of space. One day you'll be acting like 1 TB is normal for a computer. Why? Because we keep pushing it up subtly. There was a time when 100 megs was an enormous drive, and Doom taking up 20 megs for data was considered criminal (though worth it). But we wouldn't need bigger drives then would we?

Why do we need a gig for gmail? (because we don't want to ever have to delete anything if we want).

This doesn't even go into video and visual artists who can eat up hard drive space extremely fast.

Re:product looking for a market

[cliffski]

My whole point is that 2) is not true. Or if it is, they keep quiet about it. Have you ever seen Mean Time Between Failure rated alongside capacity in any consumer hard drive marketing?

Your argument seems to be "more is always better". That's not always true. a bigger engine isn't always a better engine. beyond the point where it gets you from a to b, its often just wasted effort, expense, and resources.

Given a choice of a high-reliability, quiet, high performance 200 gig drive, or an 800 gig one without those features, I'd choose the smaller drive, and I suspect many others would too. That's what my post was saying. Chill out.

Re:product looking for a market

[EvilNTUser]

Seagate offers five year warranties, and companies have in fact recognized that people want even better reliability. Wait a few years and you'll be able to replace your smallest drives with flash.

I never said more is better. I said more is better for people who need more.

Re:product looking for a market

[businessnerd]

Ok so as the posts above me illustrate, yes there is a need for more and more storage. Datacenters are probably the biggest reason. However, I'm right with you on the home PC market. Yes there are the oddballs the do need lots of storage. In fact, I recently became one of them after I built a MythTV box. It really sucks when you only get half of the movie because the hard drive filled up. But on my regular desktop, I've got 40GB and don't plan on expanding that anytime soon. I used to store all of my mp3's on there until I built the MythTV box, but even then, 40GB was plenty.

My real complaint isn't so much that I don't need it, or you don't need it. My complaint is that Dell, HP, etc. are telling Joe Sixpack that he DOES need it. Call up Dell and order a PC. They are going to ask you some questions in order to build a PC that "fits your needs" (read "fits Dell's need to make money). They will ask what you are going to use the computer for and Joe Sixpack will say "for the intarwebs and that e-mail thingy." Dell will then say "Oh well you're going to need some extra storage. We can upgrade you to 160GB for only $50!" Joe Sixpack doesn't know better so he goes along with it and never uses more than 10GB ever. The idea that more is better is not true if you don't need it. I normally don't recommend any more than 80GB for anybody unless they have a very special need for it (i.e. video editing, massive image archives, way too into porn) and even still that seems to be pushing it.

Re:product looking for a market

[tbischel]

wow,you sound like my dad in the mid 80's. "how the hell are we ever gonna fill an eight megabyte hard drive?!?"

Re:product looking for a market

[Kjella]

Well, for one laptops have taken a major upswing, which means smaller (2.5" or 1.8" HDDs), businesses produce/mine data like never before, games often take a ~5GB install these days, digicam pics are quite a few GB, mp3s, HTPCs, DV footage is huge and so on so I don't think HDDs are in any danger. But yes, I think it's slowed down because on 3.5" HDDs the premium is rarely worth it - the people you talk of buy one drive, you buy four and RAID5 them. You don't pay the premium of a top-capacity disk.

Re:product looking for a market

[wolfgang_spangler]

Why innovate on capacity? it's the one major metric that most people have stopped caring about. I'm not being a luddite, for a long time disk capacity *was* a major issue, and we regularly ran out of space. I think that time is over. Because the only use for computers is surfing the web and playing games right? That is the only thing they could possibly be used for. And hey, if *you* don't need more than 80GB then nobody else does right?

If you feel a sudden lurch in your mind..that is you finally understanding that the world doesn't revolve around you.

What a crazy narrow view you have. Do you really think that because you and your friend don't use a computer for anything useful that others don't?

I have 100TB in my datacenter, and I need tons more space. I can't even have all of my data online at one time due to space limitations...but as long as you can play your computer games and check your e-mail I guess everything is ok huh...

Re:product looking for a market

[What'sInAName]

Well, I don't know if I can say that I *need* more disk space, but I could sure find a way to make use of it! My home setup includes a MythTV DVR setup, and my workstation/home server has a 500Gb drive which is fast filling up, because I'm working on putting all of my DVDs (converted to xVid) on it. My media partition is exported to my DVR, so then I can just pick what I want to watch from a menu. None of that messy getting up from my comfy chair to search for a disk.

Video is what is going to drive this for the consumer market, if anything...

Re:product looking for a market

[asuffield]

Datacenters don't necessarily want larger disks. Frequently, they are performance oriented and are more interested in spreading their dataset across a larger number of spindles for increased performance.

Higher data density == faster. There are real, serious physical limits in the speed at which a hard drive can spin and still be reliable. That's why your desktop drives are still spinning at the same speed they were 10 years ago - we can push them a little faster and still be reliable, but the cost goes up significantly, because the increased mechanical stress requires much tighter production tolerances. To get data through the head faster, since you cannot really alter the rate at which disk surface passes under a head, you have to pack more data into that surface.

There was a time when the answer was things like "use more heads". We did all those things. Now we must pack more data into the same physical area, if we want to continue improving performance at the rate we are used to.

Coincidentally, boosting the data density is how you increase the capacity of the disk.

So yes, datacenters want larger disks. Datacenter managers may not be aware of this, but it's true.

(At any given technology level, a manufacturer may have larger+slower models and smaller+faster models - that's irrelevant to the subject here. The process of research, which this article is about, will have to drive both performance and capacity up in tandem - it's not an either/or thing)

Re:product looking for a market

[cliffski]

did you actually READ my post? I'm talking about the VAST majority of users who use their home PC to check the news headlines and chat on msn. has it ever occurred to you that all the slashdot posters talking about how they are building data centers for particle accelerators just might *not* be typical computer users.
I made it quite clear that I was referring to joe average in my post, and *still* got an army of people pointing out business uses for it, usually with some sarcasm and abuse added in for free.
Ho hum.

Re:product looking for a market

[Lord_Dweomer]

Hi,

I'm the PC gamer market who buys/downloads lots of games that take up several gigs each. I also pirate a LOT of music, TV, anime and movies (many MANY gigs of which are pr0n even though I won't typically admit it). I may also do occasional video editing work making music videos and youtube videos for my friends. I need this drive, and am willing to spend good money on it right along with the several grand I dropped on my new .

Yours truly,
Joe Average Gamer

In all seriousness...without citing some studies I'm having a hard time believing your claim. While I think the masses certainly don't use that much space I know the more advanced users (who are willing to spend more on these sorts of things) certainly do. I fall squarely into that category and I filled up 2 150GB HDDs in the first month I had them. I just bought myself an external 200GB drive and am already half full on that.

The other thing you're not taking into account is HD quality video. When people start downloading and pirating that, it WILL need that kind of capacity.

Re:product looking for a market

[ivan256]

Higher data density produces marginal performance increases, but only if the capacity increasing technology works at the same angular velocity as the previous technologies. Adding an additional spindle can significantly increase performance (relative to the usual 12 to 20% bump in capacity). Regardless, these high densities are not finding their way into the 10k and 15k rpm drives yet, because vendors can't produce high RPM disks at these capacities with the 4,000,000 hour MTBF that many datacenters demand and the vibrational tolerance that all datacenters require. Many high capacity disks actually perform significantly worse in a datacenter or array setting than they do in a workstation because the vibrations associated with the environment cause the heads to lose the track on the platter more frequently than occurs otherwise. By significant, I mean 75% reduction in throughput in an array compared to running the drive by itself. This phenomenon can cause larger capacity drive to perform worse than the smaller drives even though they should be faster in theory. Also keep in mind that while the trend in increasing data density is making its way into datacenters, this is mostly in the form of reduction in physical device size, and not in a capacity increase. Seagate in particular (who this article is about) is moving their enterprise roadmap in the 2.5" direction.

Notice that in both of my comments I carefully chose between "many" and "all" when I was making statements. I'm also not considering anything with a single array to be a "datacenter".

Re:product looking for a market

[cptgrudge]

I care, and I'll bet others still do as well.

I have about 15GB of ripped music from CDs. I have 20GB of ripped movies from DVDs. I have 40GB of ripped TV shows from DVDs. That's 75GB of media alone. Now, legality aside, you add torrent sites like thepiratebay to it all, and you definitely need that space. I know people that have 200GB or more of downloaded media files.

Then again, I watch most everything through XBMC, so my use may be uncommon.

Re:product looking for a market

[AK Marc]

In my experience, PC hard disks are already way too big. A friend of mine uses his 100 gig drive for some emailing, websurfing, playing a few games, and music playback. Last time I checked his PC it was over 85% empty. And most of the space that was consumed was the O/S.

It seems that 10% will only use 20 GB or less and 10% would fill a 100 PB drive, if you gave them a fast enough Internet connection and a Netflix account. This isn't for the people that get one email a week and go online once a month to look up movie times.

Re:product looking for a market

[Chandon Seldon]

Comments like this have been stupid since I first heard them in 1993.

Come back with me for a moment to 1998. A four gig hard drive was pretty nice... it could hold a Half Life install and a StarCraft install at the same time. And people said "Who needs a six gig drive? Most people don't, that's for sure."

Here's how this works: High end stuff is adopted by people who want it. That doesn't need to be you. As time passes, the high end becomes the low end and no-one wants a computer without at least a crappy old 40 gigabyte hard disk. If all you really do is surf the web, check email, and write Word Processor documents, I've got a 600 meg hard drive that will work fine for you. The fact is that you want to be able to do other stuff too, and as time progresses that "other stuff" will eventually want a 37.5 terabit hard disk.

Re:product looking for a market

[cliffski]

"If all you really do is surf the web, check email, and write Word Processor documents"

you just described what 90% of normal computer users do with their PCs. Just because all your friends have streaming media servers and RAID arrays doesn't mean that's representative of everybody else.

Re:product looking for a market

[Chandon Seldon]

There's a big range between the users who really only need the 600 meg hard disk (who do exist) and the users who are streaming high def video off their raid array (those guys exist too). Your claim that 90% of users fall in the former category doesn't seem to be accurate.

Most computer users I know - and very few of them are early adopter / high end type users - have at least a few audio or video files on their computer. Some of them have iPods and use iTunes, which takes disk space. The kids play video games, which take disk space.

I absolutely agree that there are a large number of users who wouldn't notice the difference between a 60 gig hard disk and a 600 gig disk today. That doesn't mean that 90% of users would be fine with 6 gig drives... in fact, I bet that 90% of users would feel constrained by that.

Re:product looking for a market

[couchslug]

Ought to be modded "(Score 5, Accurate)".

Re:product looking for a market

[Threni]

> wow,you sound like my dad in the mid 80's.

I am your dad. Defragment your hard drive - it's a real mess.

Re:product looking for a market

[drsmithy]

Datacenters don't necessarily want larger disks. Frequently, they are performance oriented and are more interested in spreading their dataset across a larger number of spindles for increased performance. They end up using terabytes of capacity for gigabytes of data. Seagate in particular has shifted their roadmap from capacity to performance in their enterprise and business class products. High capacity is reserved mostly for end users.

Datacentres are making more and more use of these huge, "consumer level" drives in "nearline" storage arrays that are used to hold less frequently accessed data.

Datacentres most assuredly *do* want larger disks, it's just the manufacturers aren't delivering them in the 10k and 15k RPM packing they'd like to.

Re:product looking for a market

[drsmithy]

Ok, so on the more general point of high capacity 3.5 inch drives, Does anyone really need these? In my experience, PC hard disks are already way too big. A friend of mine uses his 100 gig drive for some emailing, websurfing, playing a few games, and music playback. Last time I checked his PC it was over 85% empty. And most of the space that was consumed was the O/S.

I'm predicting an explosion in HTPC "back ends" drive by the Xbox 360 and Apple's "iTV" thingamabob over the next 1 - 2 years. Once people get discover they can put all their "media" into a single place with quick, easy and instant access, they'll go crazy for it.

Heck, I've already got 4TB (raw - about 3.2TB usable - 16*250GB) in my server at home, and between DVDs, TV recordings, music and software, it's only got a few hundred gigs free. I expect to fill it before the middle of the year and extend with another 5 or 6 500G drives then.

I'll get a new PC next year for vista (I need it for checking games compatibility) and no doubt it will come with a 500-1000GB drive as standard. I'd rather it didn't, I've got by for years with my 80gig friend here. If theyt *really* want to innovate on disks innovate here:

That 500G drive will be substantially faster than your 80G drive, if nothing else.

Power consumption (esp with electricity prices going menatl as they ahve in the UK)
Seek Time
Cost

Power consumption from hard disks is not much - practically zero compared to the CPU, video card and monitor. Increasing data density *does* improve seek times, and the only other way is to increase rotational speed, which subsequently drives up power consumption, heat output, noise levels and cost (due to tighter manufacturing tolerances). As for price... Drives are already dirt cheap anyway - I think I saw a coupon on dealnews only last week selling 500G drives for US$130.

Capacity is increasing because that's what is driving the industry. Less so at the typical consumer level, I'll agree, but business is generating mind-boggling amounts of data and they want to keep more and more of it online.

Re:product looking for a market

[drsmithy]

It's still a major issue for me. You're right, I'm not an average joe when it comes to storage needs, but does that mean that nobody should produce a product that fills my need? My 1.2 Terabyte RAID array is full, and I am currently wondering how the hell to add more storage and migrate the data without simply building a whole new machine.

Shouldn't be *that* difficult. What's your existing configuration ?

Re:product looking for a market

[ivan256]

Datacentres are making more and more use of these huge, "consumer level" drives in "nearline" storage arrays that are used to hold less frequently accessed data.

And drive vendors have responded with "nearline" product lines to resolve the problems that occurred (and that the customer didn't expect) when customers used consumer grade drives in environments they weren't speced for.

There are always customers that want as much capacity as possible. All I'm saying is that the vast majority of the customers who have high capacity requirements without high performance and/or high reliability are individual consumers.

Re:product looking for a market

[Zenaku]

My media center PC has 4 400GB drives in a RAID-5 array. It also has 2 other 200 GB drives in the machine. The RAID array is now full. As there is no room to add more drives, and I have no other machine capable of hosting the RAID (The drives are all SATA, my other machines are older and can only handle 2 SATA drives at most), my only option seems to be replacing the drives one by one with bigger ones, and rebuilding the RAID from the other 3 each time. This is undesireable, however, since I would just end up with 4 new drives that each have a 400 GB partition on them for the raid, and leftover space. I could then make a second raid 5 across the remaining space, but. . . that's not at all what I want.

So it seems like in order to have a new bigger RAID and the existing running at the same time (in order to migrate the data) I'd need two machines capable of hosting 4-drive RAID 5 arrays.

Seagate reliability?

[michaelvkim]

Gigantic hard drives are great and all, but I'm especially wary of anything Seagate releases that's new.

My first large hard drive was a Seagate 120GB 7200.7 that still works to this day. It's one of my favorite drives and has never let me down.

I needed more space so I buy the then top-of-the-line Seagate 300GB 7200.8. I believe this was the first to use Perpendicular Recording Technology. I backed up all of my precious data on there and went about my business, only to realize that after 8 short months, the drive had completely crashed and took with it all of my data. Slaving the drive did not work, no program I used to recover lost files could detect the hard drive... it simply disappeared from Windows and was never seen again.

There are lots of similar stories if you just do some online searching. Since this isn't just a localized case, I'm justifiably wary of any new technology that Seagate releases. Everytime Seagate implements a new technology in their hard drives, I make sure to wait a few generations before buying it. This way, the price is lower, bugs are fixed, and hopefully I'll be able to keep my data for longer than a few months.

Re:Seagate reliability?

[blackicye]

I'm currently running 6x Seagate 300GB 7200.9s in RAID 5.
and 2x Seagate 320GB 7200.10s in RAID 0, both arrays are
running on a PCIE Highpoint 2320 8 port controller.

I've not had any problems at all so far *knock on wood*

the 7200.9s are made in Singapore, the 7200.10s in Thailand,
I heard a few reports of issues with the earlier batches
of 7200.10s that were made in China (there were supposed to be
issues with the motors), but I've never owned one of those models.

The only times I've had problems with any particular model of
hdds were during the IBM Deathstar 75GXPs saga, sporadic batches of
Seagate, WD and Maxtor 160GB SATA1 Drives and most recently with
80GB and 160GB Samsung SATA hdds.

Re:Seagate reliability?

[afidel]

The 7200.8 does appear to have issues. The Storagereview reliability database has it ranked in the 31st percentile reliability-wise, although the limited number of entries (only 220) might be skewing the results a bit. The 7200.7 on the other hand is in the 89th percentile with nearly 800 entries. The majority of Segate products listed in the database with a statistically significant number of entries are ranked in the 90+ percentile for reliability. Personally I have installed about 300 Segate HDD's in servers in the last 6 months and have only had one failure, but these are enterprise SCSI and FibreChannel drives.

Re:Seagate reliability?

[Fweeky]

Everything dies more often at the front and back end of the bathtub curve, and Seagate are the world's most popular disk manufacturer, so yes, there will be plenty of "my Seagate killed my dog and committed suicide after only 13 microseconds of use!!111!11" anecdotes. That's all they are -- anecdotes. Unless you've got a significant sample size, you can draw no conclusions, though if it happens a lot you might think about using a different courier.

Of course new technology often does have a higher than expected failure rate, at least to start with. It's not unique to Seagate.

Re:Seagate reliability?

[DamnStupidElf]

My first large hard drive was a Seagate 120GB 7200.7 that still works to this day. It's one of my favorite drives and has never let me down.

Someday it will fail. What are you planning to do about it?

There are lots of similar stories if you just do some online searching. Since this isn't just a localized case, I'm justifiably wary of any new technology that Seagate releases. Everytime Seagate implements a new technology in their hard drives, I make sure to wait a few generations before buying it. This way, the price is lower, bugs are fixed, and hopefully I'll be able to keep my data for longer than a few months.

I would expect that there are (failure_rate)*(drives_purchased) stories out there, all true. What made you think that backups were unnecessary?

Re:Seagate reliability?

[smoker2]

I've got 2 Seagates running in my media server right now. 1 x ST3200822AS and 1 x ST3250823AS

The 200GB disk has been in there nearly 2 years, and the 250GB about a year. I run them in a LVM2 setup to make 1 large volume, and so far, I've not had a single issue with either. I don't intend to leave them for ever, as they fill up (close right now) I have another 300GB drive ready to go in, which I will then migrate the data onto from the oldest drive. That drive will then be used for something else. As drive capacities grow, my storage grows too, without having to limit myself to 1 or 2 disks.

If I could find a reliable consumer SATA II controller card, I would have another 3 disks in there.

BTW, they don't get used for swap or anything intensive apart from streaming video over the network, so maybe that's why they've lasted.

Incidentally, the OS is installed on another Seagate - ST340014A - an old 40GB that has been running for over 3 years without incident.

Re:Seagate reliability?

[mako1138]

I needed more space so I buy the then top-of-the-line Seagate 300GB 7200.8. I believe this was the first to use Perpendicular Recording Technology.

Nope, the first Perpendicular Seagate desktop drive was the 7200.10.

Imaging medium deluxe

[butterberg]

I don't even manage it to completely fill my current 120GB HD. But, provided I manage it some day, such a 37.5 TB beast would serve me well to create an image of my HD every day of the year. ;-) Unbelievable!

Does anybody know, how large the current index of Google is? Would such a huge HD be enough for storing it?

Easy answer

[SNR monkey]

Hi-Def pr0n

Adult entertainment always spurs innovation.

Re:Reliability?

[Sneakernets]

Well, if it doesn't work, just use a bigger HAMR...
Tim Taylor would be proud. :)

Re:HARM = Anti Radar

[cowscows]

It always seemed sensible to me to assume that they just sort of added "High Speed" to that to make the acronym cooler. I would imagine that while some missiles surely move faster than others, they all tend to move at what someone would consider to be a high speed.

Re:37.5TB HDD

[archen]

Assuming you have two drives and one is twice the density as the first and both are at 50% capacity, then assuming the data is on the first half, the head should have to travel half the distance where the density is twice as high. That should affect latency as well as twice as much data will pass under the head per revolution.

I'm pulling that explanation out of my ass though :)

Jaron Lanier approves

[gelfling]

He said that it should be possible in the near future to contain the entirety of a human brain on hard drives and be able to download your 'self' to them. Maybe we're close.

Re:Jaron Lanier approves

[jrwr00]

Then what would really suck is when the hardrive crashed to hell, now that WOULD be killer :)

Re:Jaron Lanier approves

[KokorHekkus]

Storage space wise we seem to be getting there a couple of hd generations even if were not there yet (in terms of affordability). Brain has around 10^11 neurons (one terabyte is 10^12 bytes, SI standard). Each neuron can make connections to thousands or ten thousands other ones. Even with some clever mapping of connections I'm guessing we would need something in the petabyte scale. (Source for number of neurons and connections: http://www.newscientist.com/channel/health/brain/d n9969)

Then of course there's the question if having a brain mapped out on something as relativly slow as a hd cluster could be useful.

Re:Jaron Lanier approves

[KokorHekkus]

I do belive that the speed of the hd cluster would indeed matter. I think you underestiamted the amount of data that needs to be manipulated. Let's guesstimate that only 1% of the neurons are activly used in the average case and only those needs to be changed in the model. An example: per 1 TB discspace, with 1% neuron utilization and a 6 Hz functional frequency you would need to at least access 60 GB per second .

Re:Jaron Lanier approves

[Hoi Polloi]

I'm guessing we would need something in the petabyte scale.
Good luck querying that database!

ANOTHER LIE

[Anonymous Coward]

This drive will not be 1TB. It's another scam. Rather than actually be a 1GB drive, as in 1,099,511,627,776 bytes it's a 931.32~ GB drive as in 1,000,000,000,000 bytes. Yep, 69GB short of a Terabyte. It's just falsely advertised as a 1TB drive.

Hard drive makers:
Kilobyte = 1024 bytes
Megabyte = 1024 kilobytes
Gigabyte = 1024 megabytes
Terabyte = 1024 gigabytes

Label your fscking drives accurately.

Re:ANOTHER LIE

[TheThiefMaster]

Better yet I have two differently sized Maxtor DiamondMax9 160GB hard-disks, with one defining 1 GB = 1,000,000 kB (with 1kB = 1024 bytes) and the other 1GB = 1,000,000,000 bytes.
Works out as ~152.6GB for one and ~149.0GB for the other. More than 3.5GB difference between two drives labelled with the same size. Suffice to say it annoyed my raid controller when I tried to add a 149GB disk to an array made from 152GB disks.

Re:ANOTHER LIE

[tomee]

Sorry, but this bothers me. Kilo, Mega, Giga and so on are and clearly defined as factors of 1000. Why should this be any different for computers? And then only for sizes, but not for instance for bandwidth? I personally would like to see these binary versions of Kilo, Mega etc. go away completely for the sake of clarity and standards.

Re:ANOTHER LIE

[Junior J. Junior III]

Eh, close enough. Who's really going to miss that missing 69GB, anyway? That's barely, like a small shelf of DVD movies. A pittance barely worth mentioning. I routinely lose 69GB in the wash. Never underestimate the bandwidth of a guy wearing sneakers and cargo pants with a buttload of flash drives!

Re:ANOTHER LIE

[nsanders]

This is not a lie, the submitter labeled it wrong. As the article says, "This technology, however, is expected to peak out at about 1 terabit per square inch." As you can plainly see, they said BIT. Only the submitter has it wrong.

Re:ANOTHER LIE

[Soul-Burn666]

http://en.wikipedia.org/wiki/Binary_prefix

Kilobyte = 1000 bytes
Megabyte = 1000 kilobytes
Gigabyte = 1000 megabytes
Terabyte = 1000 gigabytes
Kibibyte = 1024 bytes
Mebibyte = 1024 kibibytes
Gibibyte = 1024 mebibytes
Tebibyte = 1024 gibibytes

Re:ANOTHER LIE

[qazsedcft]

No. The definition of "gigabyte", etc. has been officially changed. See http://en.wikipedia.org/wiki/Binary_prefix

Re:ANOTHER LIE

[benzapp]

Is a kilometer 1024 meters?
Is a kilogram 1024 grams?

It is the software makers who do not understand these historic terms. Fight the redefining of words!

Re:ANOTHER LIE

[Mirar]

Sorry, it's not. There is no reason to use the 2^10 or 1024 step for computers, other for nice figures when counting something that needs to be in 2^n sizes (ie, storage limited to a bus width, like RAM) - disk space, file size and bandwidth is not limited to 2^n, so using normal prefixes makes sense. (References, see other posts.)

Did you hear that, file tool makers? Stop counting in 1024^n! It doesn't make sense!

Re:ANOTHER LIE

[phasm42]

It's not any different. Kilo, Mega, Giga etc are factors of 10... in binary. Factors of 10000000000 to be exact. Fundamentally, memory is addressed in binary, so storage related to memory will come out to a number related to a power of two. When you buy a 64MB stick of RAM, it means 64 * 2^20, which is simply 100000000000000000000000000 in binary (or 4000000 in hex). By using powers of two for the units, you know precisely, to the last byte, how much memory that is. If you were to write it in decimal, that's 67108864 bytes. Saying 67MB would simply be an estimate.

Re:ANOTHER LIE

[phasm42]

I'm guessing that you can't read binary. Hint: 10 in binary = 2 in decimal. 10000000000 in binary = 1024 in decimal. These units originated from working with binary numbers -- everyone either understood or was unaware that the units were based on powers of two until some hard drive makers decided to inflate their numbers by using powers of ten. Since software displaying capacity uses powers of two for their units, the numbers didn't match up and that's when people noticed.

If those marketing assholes had just stuck with powers of two like every one else, all would be well. Hell, using powers of two skews the numbers in the customer's favor -- 67,108,864 bytes is called 64MB, which is smaller than the inflated 67MB figure a marketeer would use. If all the units are based on powers of two, then everything works well together and these weird inconsistencies won't be present.

Re:ANOTHER LIE

[CKW]

.

Fuck the IEC, they can go to hell. No way am I *ever* using these retarded "bi" words.

Geee, never heard of the IEC before either. And what's this word - "Electrotechnical".

Ooooh, I betcha I know. Let me guess. A front for the very same assholes who want to fuck with the 20-30 year old "kilo/mega/giga" prefixes for their own damn benefit, so the electronics they sell can say "kilo/megal/giga" without actually BEING a power of two - exactly as everyone in the COMPUTER/SOFTWARE world has been using the terms for 30 years.

They can go fuck themselves. The only people who ever had problems where the ones who insisted on using the SI definition for computer jargon.

( Thing is about words in language - they have to be easy to pronounce and remember in that language. You try and pronounce these words 10 times really fast. The repeated b's get stuck on the tounge and they sound retarded to boot. )

.

Re:ANOTHER LIE

[Arcaeris]

While the names may have been officially changed in 1999 and many of us having been working on computers for far longer than that, that's just old habits dying hard - the real problem is one of a perception.

Windows (and I assume Mac OS?) continues to display file size in terms of base 2, and HD manufacturers have bought into this base 10 thing (to make their hard drives sound larger).

I don't care either way which one they use, as long as both groups agree on the same thing. This discrepancy between what is on the HD box/computer website and what is shown in the OS confuses a lot of people, and it's a pain to explain to the average Joe consumer why he isn't getting what he thought he'd be getting.

Re:ANOTHER LIE

[CKW]

.

Jeezues H C. The proposed pronunciation is even more fucked up than I'd imagined.

What we all know and love as KB/kilobytes (easy to type KB, easy to say kilobytes) is now supposed to be KiB (shift, lowercase, shift - pain in the ass) and ki-bee-bytes or ki-bee-bits.

ki-bee-bytes? What is that, dog food?

Switching to a new system now after 40 years - whose new acronyms and names and pronunciations were choosen by morons -- would create massive amounts of confusion.

.

Re:ANOTHER LIE

[Treacle Treatment]

I've always resisted the ki-bee-byte and friends designations because they are so hard to pronounce.

Re:ANOTHER LIE

[D4rk Fx]

How big are hard disk sectors? That's right. 512 bytes. An average cluster is 8 times that, or 4096 bytes.

It's still a completely valid, and necessary way of describing file sizes. If a hard disk manufacturer swants to make the sector sizes 500 bytes instead of 512, then by all means, the Original SI designation of kilo, etc. would be a better way to represent the total space. But since they still use 512 bytes for a sector, it should always be in 2^n.

Re:ANOTHER LIE

[Staz]

Yesterday, some friends and I, decided to form the International Committee Of Terms To Be Used On The Internet, ICOTTBUOSTI, and we decided to change it back again. Now, abide our mighty decision

Re:ANOTHER LIE

[Deadplant]

The IEC can go take a long walk down a short pier.

I'll never use their ridiculous naming scheme.
It is a stain on the language.

Now I'm going to go try and clear up a few gibblebytes from my tribblebyte drive array.

Re:ANOTHER LIE

[AK Marc]

At least for English, a standards body doesn't define words. That's why HD manufactureres now, and until the end of time, put "gigabits = 1000000000 bits" somewhere on the package of ever hard drive (or whatever mega/giga/peta bits/bytes fits their drive size). If it was truly a real standard, then there would be no fear of lawsuits that force the explanation of which GB is used, 1024 or 1000. So, based off law and common usage, for all things computer related, the use of GB is referencing the 1024 standard.

The standards body should have recognized that SI prefixes in front of "bits" or "bytes" implies the base-2 standard and written the standards the other way.

Kilobyte = 1024 bytes
Megabyte = 1024 kilobytes
Gigabyte = 1024 megabytes
Terabyte = 1024 gigabytes
Ki10byte = 1000 bytes
Me10byte = 1000 ki10bytes
Gi10byte = 1000 me10bytes
Te10byte = 1000 gi10bytes

There, all fixed. (Pronounced ki-ten-byte, me-ten-byte, etc.) Yeah, it's stupid and pointless, but it's better than what they did come up with.

Re:ANOTHER LIE

[toddestan]

Is a kilometer 1024 meters?
Is a kilogram 1024 grams?

It is the software makers who do not understand these historic terms. Fight the redefining of words!

Historically,
1 kilobyte was 1024 bytes
1 megabyte was 1024 kilobytes

Sure, this flew in the face of how the kilo- and mega- prefixes were used elsewhere, but that was how it was done. It wasn't until the harddrive manufacturers decided to redefine the terms so they could sell their 95.37MB drive as a "100MB drive" and mislead customers did all this confusion start.

Re:ANOTHER LIE

[theLOUDroom]

The question is:

What right does the IEC have to redefine terms that were in common use 20 years before this standard existed?

A bit is not a part of the CGS system of units. The terms were already well defined before this "standard" existed and this is nothing but a dishonest attempt to redfine the terms for the benefit of the hard drive manufacturers.

How about some fucking common sense people? Computers work in base TWO.... why the FUCK are you going to measure storage in base TEN? Could you imagine if we allowed coporate marketing departments to redfine ALL our technical terms? Grade 8 bolts would suddenly have the same composition as drywall screws.... people would fucking die because us poor egineers could not keep it straight. But hey, we'd all be following standards....

Re:ANOTHER LIE

[ElephanTS]

No, that's a kibimeter and a kibigram respectively.

Someone was going to say it . . . (and it had to be me :-p)

Re:ANOTHER LIE

[dangitman]

What right does the IEC have to redefine terms that were in common use 20 years before this standard existed?

What right do computer geeks have to redefine terms that were in common use hundreds of years before computers existed? Kilo has always meant 1000, not 1024.

Computers work in base TWO.... why the FUCK are you going to measure storage in base TEN? Could you imagine if we allowed coporate marketing departments to redfine ALL our technical terms?

You can measure in whatever base you want - but why go and fuck with perfectly consistent english and scientific notation to define your own strange exception? Perhaps the people working with computers should have come up with a word to mean "1024" rather than perverting the existing scientific standards?

Re:ANOTHER LIE

[Mirar]

That doesn't mean that the hard disk is an even 2^n. Just an even n*512, or depending on your filesystem, n*4096. Where n is very, very large; extremely much larger than 512.

And I can still store four files of 72, 16, 48 and 19 bytes in that sector. The basic storage unit doesn't mean I can't use it partially. Hence, file sizes of even n*512 doesn't make sense either, unless the file system actually stops me from using smaller units.

Re:ANOTHER LIE

[Mirar]

I think I should offer a better explanation.

I have a hard disk here, that has a data storage data of 476993802240 bytes.

Using gibi bytes, it's 444.235107 GiB. This is not an even figure, even though it has 1024 as a base.

Hence there is no reason to use (2^10)^n prefixes. You might just as well use the metric prefix, 476.993802 GB.

The only reason to use the (2^10)^n prefix is that file system tools lie and give you the wrong numbers, so you get confused when you buy a 480GB disk and can't fit more than 445GiB on it, claimed by the file system tools to be 480GB. The disk manufacturers are not the ones that bend the truth here.

Since we decided to stick to using 10 as base, I think we should stick to using 10 as base. (I know this seems very hard for some regions who use 12 and 1/8ths and other numbers as base for some calculations (inch-foot and bolt sizes, for instance), but that's just another thing those regions have to overcome and does not help to settle the confusion.)

If we shouldn't stick to 10, I suggest we go for 60 or 16. 60 is a nice number (as discovered early in history) and you could do really quick calculations if you know the full multiplication table - but 16 is an ok number as well, the multiplication table is big enough, and it's very compatible with computers. ...but lets stick to 10 for now, ok?

Re:ANOTHER LIE

[Mirar]

s/to be 480GB\./to be 445GB, or even worse, 457GB or 468GB.../

Re:ANOTHER LIE

[theLOUDroom]

Those arguments would make sense if bits and bytes were somehow related to the CGS system of units. They aren't.

Base two is automatically implied by using the terms "bit" or "byte".
If they want something that functions differently, they are free to invent new root words which take on different meanings with those same modifiers.

Re:ANOTHER LIE

[dangitman]

Base two is automatically implied by using the terms "bit" or "byte".

Exactly. So why the need for this 1024 shit? You're already talking about units in multiples of what's "native" to the machine/software.

Re:ANOTHER LIE

[theLOUDroom]

Exactly. So why the need for this 1024 shit?

Because 2^10 = 1024, not 1000.

You're already talking about units in multiples of what's "native" to the machine/software.

Which is my point. Addressing of data is done in base two, not base ten.

The switch to base ten, was a marketing tactic of the hard drive manufacturers. It is dishonest and deliberately misleading.
If you look at other things like RAM, a 1Gb RAM chip gives you 1073741824 bits (2^30).
It's that way today, and it was that way twenty years ago.

1 Kb = 1000 bits is the brainchild of sleazeball marketers.

Re:ANOTHER LIE

[dangitman]

Because 2^10 = 1024, not 1000.

So what? "Kilo" does not refer to 2^10. It refers to 1000, and is a base 10 modifier.

Which is my point. Addressing of data is done in base two, not base ten.

So what? How does accurately using "kilo" prevent you from addressing data in base two?

The switch to base ten, was a marketing tactic of the hard drive manufacturers. It is dishonest and deliberately misleading.

So, then why do other standards, like network throughput, refer to kilo as 1000, and not 1024? And if the industry had used the prefix correctly in the first place, then HD manufacturers would not be able to get away with this "trick." The industry has only itself to blame for muddying the waters and allowing this to happen.

It's that way today, and it was that way twenty years ago.

That's actually a 1 GiB chip, not a 1GB chip.

Coming Soon: The LTO-48!

[Penguinisto]

...the tape will be in a cartridge that holds a spool 65cm in diameter, holds approximately 600TB (1200TB w/ compression) and will require an autoloader that eats at least one rack for the entry-level 8-tape kit. /dev/nst0 will weigh in at 38kg, and cleaning will require a tape w/ 6000-grit sandpaper in place of media.

All BS aside: you do bring up an excellent point. I'm a guy who has to do backup/recovery, and I've found that even a fully compressed LTO-3 will barely --just barely-- hold up to 1.2TB if you rig it right (by combining hardware/software compression, and the love that Bacula gives it (though admittedly sparse file handling most likely has inflated the reported amount of stuff).

Anyrate, that boils down to --maybe-- two full HDD's if the two are 500GB SATAs.

The good news is, after you pare down the crap you really don't need to backup, it usually isn't all that much for most companies. You can safely exclude out most of the OS itself for starters... w/ kickstart on RHEL and a .ks file that replicates what you've got on a given server (partitions, packages, etc), you can cut a LOT out.

Even more good news - if you get up a monster RAID array of similar drives (full SAN kitting or just attached to a big ol' server, no biggie), you can use it instead of tapes for most of your day-to-day backup. Then latch your tape drive or autoloader onto it and only commit to tape the reallly vital stuff that requires a long retention period. Most backup software suites (even Bacula) support writing to file as well as tape, so this shouldn't be too big of a problem for a sysadmin if s/he knows what s/he's doing.

Adaptation and all that.

But then, most of the servers in my care consist of a pile of RAID5'ed SCSI drives that range 36-140GB in size... and I doubt that most of them will get much bigger before it's time to replace the servers themselves. Just because you can get monster capacity on a single drive, doesn't mean that you need to or even want to.

Now if I already had a monster robotic multi-drive tape library running 24/7 now, and the boss wants to up the HDD capacity on a given pile of servers because he pretty much has to? Yeah. That would require a lot more thought and planning, and at that stage of the game a disk backup solution similar to what's been outlined above would be big and ugly, but would pretty much be what you're stuck with having to do.

...at least until they come out with the LTO-48 ;)

Re:Coming Soon: The LTO-48!

[jamstar7]

But then, most of the servers in my care consist of a pile of RAID5'ed SCSI drives that range 36-140GB in size... and I doubt that most of them will get much bigger before it's time to replace the servers themselves. Just because you can get monster capacity on a single drive, doesn't mean that you need to or even want to.

But think of the pr0n! Won't SOMEBODY think of the pr0n????

Re:Coming Soon: The LTO-48!

[Lord Apathy]

Don't worry buddy, got you covered.

Re:Coming Soon: The LTO-48!

[Just Some Guy]

I've found that even a fully compressed LTO-3 will barely --just barely-- hold up to 1.2TB if you rig it right (by combining hardware/software compression

Well, dumping compressed data to a compressed tape will actually cause it to take up more room, so what different combination of hardware/software compression are you using?

Re:Coming Soon: The LTO-48!

[LunaticTippy]

The hardware compression on my tape drives is pretty basic. I've rarely seen a compression ratio over 3 on an individual file. Software compression can compress a lot further. A text file might have a compression ratio of 70 or more. If your backup software compressed files where the software payoff is a lot bigger than the hardware one you'd end up using less space.

Re:Coming Soon: The LTO-48!

[Just Some Guy]

If your backup software compressed files where the software payoff is a lot bigger than the hardware one you'd end up using less space.

Are people using drives that can toggle compression on the fly? In my experience, it's usually on or off on a per-tape basis. A typical setup with Amanda Backup is to disable drive compression for the reasons you give, plus the fact that the backup software can know exactly how much data will fit on a tape in advance.

Funny you mention that.

[Lethyos]

The cost, longevity, performance, and capacity is completely inferior to making backups of disks onto other disks, and has been for quite some time. I have no idea why people ever stick with tape at all these days other than for nostalgia. Does it feel good to have a cartridge using a remarkably old fashion approach to data storage or are people just ill-informed?

Re:Funny you mention that.

[Penguinisto]

"I have no idea why people ever stick with tape at all these days other than for nostalgia."

...because it's easier (and far less nerve-wracking) to hand over a case full of tapes to the offsite storage courier, knowing full well that he's prolly going to just (literally) throw the thing into the back of his truck and hurry off to the next client?

True disaster recovery planning involves offsite storage of data IMHO, and tape is hella easier to transport than HDDs. Also, you don't have to worry about what order you stick tapes in, whereas with disk storage, re-assembling a RAID array would be a PITA, even with labelling.

(maybe someday tapes will be replaced with big-assed flash-storage devices? But until then, dropping a tape isn't as likely to make you wet your pants as dropping a big-arsed tape would). /P

Re:Funny you mention that.

[mcrbids]

Addressing both issues:

1) Off-site backups - that's why you back up offsite! Set up a backups-server (it doesn't need to be fast, it will be I/O bound) with a pile of cheap, big, slow HDD. Set up as RAID 5 or RAID 1 - your preference.

Then use a tool like Backup Buddy or Backup PC to back up the files.

It's automatic. It's off-site. It works with good-sized data sets. (I'm managing just over a TB these days, a number that's growing fast. I'm able to do unattended offsite backups every 24 hours, and I have several months worth of backups on file that I can access instantly)

2) Rotating thru tapes - tools like Backup Buddy and/or BackupPC take care of that, too! Assume you have 200 GB of "Enterprise Data" to backup. So you get a TB or so in your backup array, (say, 4 350 GB disks, RAID 5) and use one of the above tools. You'll see months worth of backups out of that.

Without swapping tapes, hiring anybody to take them "off site" and for just a few hundred bucks in cheap, IDE disks. Can tape do that, too?

Re:Funny you mention that.

[clydemaxwell]

Who the hell modded you up? You have obviously never been involved in a large-scale backup solution.
Disks DIE. Tapes rarely do (comparatively). Tapes, although slow and linear, are incredibly durable.

HDDs aren't exactly volatile, but they are a heck of a lot more susceptible to corruption and failure due to the fact that you have both a magnetic storage medium AND the circuitry to power and control it on one device. And if one dies, you're pretty much fucked. A tape is only one of these, and is simpler and more reliable.

So why do we do things the old-fashioned way? Because it FSCKING WORKS!!

Re:Funny you mention that.

[jelle]

"And if one dies, you're pretty much fucked."

If _one_ dies? You use disks without Raid5 or Raid01 anywhere except maybe on a laptop?

I even use raid1 and raid5 at home.

Re:Funny you mention that.

[mlts]

There is one big reason: Sysadmins won't have a heart attack if a tape clatters to the floor.

Tape has been engineered for decades for reliability. A tape cartridge doesn't have much in the way of moving parts compared to a hard disk that can go out of whack, and modern tapes like DLTs, it will take more than a clatter to the floor to make the tape unreadable.

Hard disks are great, but way too fragile for serious backups. However, I wish tape drives and tapes would come down in price like hard drives... the first tape drive with a decent price/performance ratio is a DLT-4 for $1000 or so.

To boot, there is no standard for removable hard disk cartridges for an autoloader. Yes, Iomega has the REV removable disks (great technology, can be used as WORM archival media, but expensive and not that much capacity compared to a tape drive), Imation has the Ulysses/Odyssey line, but there is no true multivendor standard for making hard disks as easy to mount and mount and swappable for robotics as tapes, with circutry designed to allow for hot-plugging and unplugging with a high number of insertion/removal cycles. What would be nice is a standard cartridge format for both 3.5" and 2.5" drives that has good shock protection, circuitry that can withstand a large number of hot plug/unplug cycles (USB isn't designed for this), uses the full SATA speed of the disk, and is easy for a company to design a robotic mechanism to label and change cartridges in a library.

Re:Funny you mention that.

[phasm42]

200GB is nothing. Many people regularly backup 5-10TB and far more. If you think everyone can get months of backups out of 1TB, you're a fool.

The off-site backup thing is irrelevant -- the same can be done with tapes.

Re:Funny you mention that.

[phasm42]

And FYI, I do use hard drives for backups here (1-2TB) because it's cheaper for our requirements. But that doesn't mean tapes are stupid and everyone should use hard drives for backup. I'm still aware of its limitations and situations where tapes will work better. Hybrid solutions are also needed sometimes, look up D2D2T (disk to disk to tape).

Re:Funny you mention that.

[clydemaxwell]

I guess I can see how you can miss that..
1) by one, I meant one of the factors (magnetic platters and electronic controller board)
2) this discussion was aimed at removeable mediums for offsite backups; I suppose you could apply a RAID across and then ship the whole RAID somewhere but it would have to be rebuilt onto your controller when you wanted to do a restore

Re:Funny you mention that.

[Rakishi]

Your solution still has the potential for all the data to get slagged through software errors (all the backups are online, all I need to do is access the backup server and zero the disks). That is unless someone swaps around disks on a regular basis but then you have the problem of disks being less durable than tape thus again making tape better.

Re:Funny you mention that.

[Beryllium Sphere(tm)]

It's not hard to find a hard disk with a non-operating shock rating of 900 g. That's before you pack it for shipping. And the magnetic material won't flake off in storage.

Re:Funny you mention that.

[LordSnooty]

I even use raid1 and raid5 at home.

Congratulations for being made of money and/or running a business from your home.

Re:Funny you mention that.

[dangitman]

And the magnetic material won't flake off in storage.

But the drive bearings will seize up if they aren't kept spinning.

Re:Funny you mention that.

[ClamIAm]

HDDs aren't exactly volatile, but they are a heck of a lot more susceptible to corruption and failure due to the fact that you have both a magnetic storage medium AND the circuitry to power and control it on one device.

Add to this 10,000 RPM platters *and* a read-write head that sits close to the platter surface and constantly skits back and forth to different areas. And when I say "close" to the platter, I mean extremely close: the drive head is held at the correct "altitude" by the air current created from platter rotation. This is a microscopic scale, folks.

Re:Funny you mention that.

[WuphonsReach]

True disaster recovery planning involves offsite storage of data IMHO, and tape is hella easier to transport than HDDs. Also, you don't have to worry about what order you stick tapes in, whereas with disk storage, re-assembling a RAID array would be a PITA, even with labelling.

The better RAID systems keep track of which drive is which in a RAID array. I know for sure that simply Linux Software RAID does it, and I'd be surprised if higher end RAID cards didn't.

Re:Funny you mention that.

[cas2000]

Congratulations for being made of money and/or running a business from your home.

i'm not made of money, nor do i run a business at home, but i run raid-1 on all my computers (linux software raid, of course). actually, there's one exception - the crappy old laptop in the back room, which only has a 10GB drive anyway.

i would run raid-5 but i can't quite justify spending the money on a hardware raid controller with non-volatile cache - without which, raid-5 is too slow to bother with.

having lost nearly all my data once due to a disk crash (and inadequate backups), i'm quite happy to spend twice as much for my disk storage for raid-1. a 320GB drive costs $AU128 today, so i have to spend $256 to get two of them. it's worth it. it's still significantly less than $1 per GB.

compared to the approx $2000 i paid for my first 20MB hard disk (a 65ms microscience clunker) back in about 1982 or 83, it's an amazing bargain.

I think it's in the Peta or Exabyte range...

[tpjunkie]

http://en.wikipedia.org/wiki/Petabyte

Wikipedia says it was estimated at a few petabytes back in '03, but now their cluster has that much RAM (!!!), so even at google, they could probably use several hundred of these.

Thank you!

[shirizaki]

Thank you Hitachi, for now I no longer have to worry about never having enough porn. With enough of these drives I should just be able to backup the internet.

Re:Thank you!

[Zaatxe]

Since when is 37.5 Tbytes enough for all the porn a man needs?

Backup Solution and a question

[WidescreenFreak]

Here you go.

http://sunsolve.sun.com/handbook_pub/Systems/C2/C2 .html
http://sunsolve.sun.com/handbook_pub/Systems/L8/L8 .html

Now, whether or not the home user will be able to afford one of the damned things is another issue :) but one of these bad boys fully loaded will back up that drive.

Although you were being a smart ass -- and I can appreciate that :) -- you do bring up an interesting question. With drives increasing so rapidly and for such inexpensive prices, you'd think that the tape drive manufacturers would be scrambling to keep up and make appropriate backup solutions more affordable for the home user. I don't mind using a mirror to keep the data redundant, but I'd still feel more comfortable having a mirror and a tape backup.

Re:Backup Solution and a question

[LibertineR]

I wasnt being a smart ass. Tape Backups have taken years from my life expectancy. Any new drive announcement triggers me to think about backing that bitch up. My clients always want the latest new hardware, with no thought as how to protect their data. If I had a time machine, I would go back and find the person who came up with the idea for DLT drives, and kill them in their crib.

Re:Backup Solution and a question

[WidescreenFreak]

:) Well, that's DLT. No wonder you're upset.

We've had incredibly good results with LTO. The one that we got (the StorEdge L8) is six times faster than DLT IV on a single tape drive and has had no media reliability problems. I really would love to get my hands on a used StorEdge L8 for home, even if it's LTO-1.

Re:Backup Solution and a question

[Penguinisto]

"Any new drive announcement triggers me to think about backing that bitch up."

I've found that counting the amount of actual data the customer/client needs stored, as opposed to the drive capacity, keeps the blood pressure nice and low. :)

"My clients always want the latest new hardware, with no thought as how to protect their data."

Part of being a sysadmin is to help guide them back into reality :) I'm kinda fortunate in that most corporate folks that I know give me the opposite problem - they want it cheap, damnit - and will only stand for purchasing just enough 'oomph' to do the job over a given period of time.

But... if you do build that time machine, can I go with you and help kill the DLT guy? We'll take a detour on the way back and get the DAT guy while we're at it.

/P

Re:Backup Solution and a question

[LibertineR]

But I am not a sysadmin. I am a consultant, usually brought in after things are too fucked up to fix on the cheap. I would love to guide clients through the build out of their data centers, but I can make a LOT more money fixing the screwups of others. When things are in the dumper, I can charge almost anything I want.

Am I greedy? Fuckin-a!

Once I am aboard, I do guide clients through most purchases, but since I work with mostly Doctors and Lawyers offices, there is always some smart guy who fancies themselves as a techie guru since they can fix hearts and livers, who wants to make enterprise-level IT decisions without consulting me, and have their IT flunkie install it.

Those are the calls I get at 3am, which will ultimately pay for my new Hummer.

Only a matter of time

[davidwr]

Here is a Capacity over time chart.

Just eyeballing the straight line, this chart shows capacity doubling every 21-22 months or so. Lately things have sped up a bit.

I don't think someone would announce a drive that was 9 years away. It looks like things are moving at a faster clip, faster even than the 18-month "Moores law" that applies to transistors.

Here is the important question on everyone's mind:

Which is doubling faster:
Drive size or the yearly porn production rate?

stupid headline

[Thaelon]

We also expect the sun to turn into a red giant within a matter of years.

Re:stupid headline

[aldheorte]

I was going to post the same thing. "A matter of years" could mean 3, it could mean 100, it could mean 'til the end of Sol, etc. Instead of what? Decades? Centuries? Millennia?

Re:37.5TB HDD

[Zenaku]

Unless I'm misunderstanding your example, that would make the seek time less, not more, as the poster I responded to was implying.

However, your example isn't quite right -- if you have two drives of the same physical dimensions (number of platters, size of platter, number of read heads, RPM) and one is twice the density of the other, and both are at 50 percent capacity with that data assumed to be on the "first half" of the drive, then the distance the arm needs to travel is the same in both cases. The arm travels half the radius of the disk.

It doesn't matter that it is passing over twice as many bits in that distance, the distance is the same. Similarly, the latency is the same. The average amount of time that the read head must wait for the data it is seeking to appear under the read head is still half a revolution. The fact that twice as many bits fly under it while it is waiting doesn't change that.

The fact that twice as many bits fly under it while it is reading the data means that the read time should be faster, but that is not seek time.

Re:HARM

[slashbob22]

You can keep your HAMRs away from my HDD as well.

Re:300 Terabits.

[arcade]

Real world KB = 1,024 bytes.
Real world MB = 1,048,576 bytes.


No.

Real world KB = 1,000 bytes
Real world KiB = 1,024 bytes
Real world MB = 1,000,000 bytes
Real World MiB = 1,048,576 bytes

Read up on the standard, boy. Heck, even ls supports --si to be correct. :-)

Time to wake up

[WidescreenFreak]

Of course, it doesn't help that I was thinking about the new $399, 1 TB hard drive that was also annouced today when I typed up my response to you, only to look up at the thread title (after submitting) and realize that you were talking about the 37.5 TB drive.

This day can't end quickly enough. :/

Some more specs

[arielCo]

From the spec sheet:

Startup current: 2.0 A on the +12V rail, plus 1.4 A @+5V

Power - Random R/W (avg): 13.6 W

(heat dissipation equals the power consumption, since there is no "output power")

Power consumption is mainly a function of the number of platters (5 in this case) which in turn determines head assembly inertia, and seek time (inversely). In any case, the inertia of one motor and hub plus five platters is surely less than four such assemblies with 2-3 platters each.

If it were simply an escalation of traditional technology, it *might* be a case of pushing-the-envelope-no-matter-what to achieve the oh-so-marketable 1TB, and the above would not apply. But nope.

As for noise: does anyone have an idea of how loud is 2.9-3.2 bels (typical) ?

Re:Some more specs

[cdrudge]

As for noise: does anyone have an idea of how loud is 2.9-3.2 bels (typical) ?

1 Bel = 10 dB. 30 dB is about the same as a quiet whisper, leaves rustling, or a typical library.

Re:Some more specs

[LunaticTippy]

Deci- is sad. So rarely used. All the other metric prefixes get used all the time but not Deci-. I guess Deka- and Hecto- aren't very popular either. They should start a cult.

Re:37.5TB HDD

[bradsenff]

You fail at math.

If both disks are 50% full, and one is 2x the density, the heads only need to travel 25% of the disk.

Re:Call it The Big Johnson

[Mr Pippin]

No need to worry. This will be addressed within 1000 years, anyway.

Professor Farnsworth: Fifty-three years old! Oooh, now I'll need a fake ID to rent ultraporn.

HAMR not HARM

[cheese-cube]

It's HAMR not HARM. Heat-Assisted Magnetic Recording. Here's the relevant Wikipedia article: HAMR.

Re:HAMR not HARM

[jez9999]

Obligatory link.

Re:HAMR not HARM

[rthille]

Yeah, this is HARM. :-)

Product will create the market, it always does.

[Kadin2048]

I think the market is right around the corner: high-definition TV.

The PVR market has been crippled in recent years because of market confusion, and compatibility problems (will my TiVO work with my cable box, etc.), plus competition for consumers' money by HDTVs themselves.

Once people get done buying their HDTV and paying off their credit cards, they're going to start looking at PVRs. I think that's a market that's probably going to explode in the next 5-10 years, even more than it has already. I also think you're going to see PVR functionality being built into the 'standard' cableco boxes, rather than as an upgrade. (Not that it will be free, they'll just charge everyone for it.)

High-def TV takes up a lot of space. That means if you want to have significant PVR functionality, you need to have a lot of local storage. 37.5TB, or 300Tb (aka 300,000,000Mb, if we use the 'marketing department' definitions) would be about 4,340 hours (180 days) of 19.2Mb/s HDTV. While that seems impossibly huge, I could imagine a future PVR using it as local cache: constantly downloading and storing programming based on your preferences. Add in a big HD movie library (say the contents of your local Blockbuster) and you can give the customer the impression of many simultaneous channels, even if they only have a relatively narrow pipe. (Narrow being 1 HD channel at a time, or a 20Mb pipe -- fat by today's standards, granted.)

Content always expands out to fill the available capacity. I remember when I first heard about the development of DVDs, back in the early 1990s. They seemed pretty ridiculously big then, too. Now I have stuff that I can't back up to DVDs, because it would be impractical to split it among so many discs as would be required. (Apple's Aperture doesn't even try to have a backup-to-DVD option, it's designed strictly to work with removable hard discs as backup 'Vaults.')

There was a time when people thought 20MB removable media was more than a single person would ever need, though we might look back and laugh. There's going to be a time in the future when 40TB looks the same way.

Re:Product will create the market, it always does.

[Jeff DeMaagd]

I just bought a dedicated 750GB hard drive for my PVR uses. I have my Mac record over-the-air HDTV. A single HD movie might take 10-20GB.

It's probably more than I really need, and I've gotten into the practice of culling stuff that I'll never watch again, but I like to keep plenty around. I also store video editing work, and that eats hard drive space quicker, 13GB an hour just for the source media, then there's the renders and ancillary files.

In all, I have just over 2GB in my computer right now. I'd really like to up it to 3+GB but I'm waiting for the drive prices to come down.

Hitachi reliability?

[electronerdz]

"The approach we've taken with the design of this product, and with previous generation products, is that we've purposely relaxed the areal density. The previous generation [500GB] drive was 100GB per platter; and, it was possible to have up to 160GB per platter. About 250GB per platter is the next bump on the areal density curve, but we've backed off from doing that in order to achieve higher reliability at this time."

This is why I use Hitachi... they are more interested in reliability over disk space. My girlfriends Seagate in her Apple crashed after 4 months. I've replaced several other Segates with Hitachi's because of crashes. I've only seen one of the Hitachi's fail so far, but that was over 2 years of constant running. But it was on a RAID, so no loss in data.

Re:Hitachi reliability?

[toddestan]

I was initially leary of Hitachi, as they basically took over IBM's hard disk division after it released a whole string of lemons. However, I have found that the Hitachi disks seem to be reliable, though some of the earlier ones were somewhat noisy.

On the other hand though, I have always found Seagate drives to be reliable too, though their entry level (often found in OEM computers) seemed to have pretty poor performance.

Re:Hitachi reliability?

[electronerdz]

I have always heard good things about Seagate, but have never had good luck with them. And whenever I say I prefer Hitachi, people usually try to tell me about the "5 year warranty" that Seagate's have. 5 year warranty doesn't mean crap to me when the drive dies. It's dead. I don't care about a warranty. I care about my information on it. And at the rate of growth of most data, 5 years is too long, and you need to upgrade after a couple of years.

Re:37.5TB HDD

[Achoi77]

with the advent of solid state harddrives coming out soon enough, how often will users be need all that data at their fingertips? Harddrives are being split into two tiers:

1: the fast, expensive type (solid state)

2: the slow, cheap type (conventional platter)

We've all seen the signs for years, only now has the tech finally been able to catch up. From a home use perspective, here's an easy prediction: SSD will store system data(OS, apps, that kinda stuff) and the HDD will store media (pictures, music, and movies movies movies). Don't really need fast seek times for media, after all.

Re:37.5TB HDD

[MyNymWasTaken]

You fail at logic.

If both disk are 50% full, density is irrelevant. The heads needs to travel the same distance - and it's not half the disk. Try ~70% = 1/2^(1/2).

Re:37.5TB HDD

[FirienFirien]

For "ms" read "milliseconds" not "minutes".

There's more to data loss than crashed HDs

[Vellmont]

When I asked why, he said that although he didnt want to buy another drive, he understood the importance of having a backup for his data.

Well, obviously he's not going to be protected from a failure of the drive mechanism. But his strategy isn't totally useless. By copying to a seperate partition he's protected himself from accidental erasure, and corruption of the data (though software that either corrupts it, or from a power failure).

It's really a poor mans archival mechanism. I'd argue that data corruption or unwanted erasure happens more often than drive failure.. though I do agree the guy shouldn't have chinced out and just bought 2 drives, RAID-1 them, and then figure out some proper archival method like tape, or even a removable drive.

you just know

[godsfilth]

that its gonna hurt bad when they use the acronym HARM especially when the H stands for Heat

Re:37.5TB HDD

[Zenaku]

You are correct that the average distance that must be travelled is 25% of the radius. In BOTH cases. The density doesn't matter.

However, I only claimed half to be consistent with the scenario my parent poster presented, which appeared to assume the worst case -- starting from the rest position and seeking the sector furthest from it. . . which would be half the radius.

300 teraBIT, 37.5 teraBYTE.

[freeweed]

FYI, 300 / 8 = 37.5

Sweet jesus, do you people not even read the summary anymore??

I freaking knew it!

[blankoboy]

I have been holding off on buying a pair of large capacity HDD's for the last 2 years knowing that we'd be seeing 1TB soon enough. But of course, I couldn't wait any longer and go buy a pair of 500GB cudas. Not less than 24 hours later they announce this. There is an MCP and he is a cruel one at that....sigh.

Re:HARM

[Flimzy]

The same has been said of Maxtor, Quantum, Western Digital, IBM, Hitachi, Fujitsu... you name it.

*Every* HD manufacturer has had a bad batch, a bad design, or just bad luck, every now and then.

The same is true with practically any manfuacturer of physical (or probably other) goods. Some car makers have bad years, or a design flaw makes it into a particular model.

It is virtually impossible to make any valid claims that brand x is better/worse than brand y based on a single model/style of product. Even if you have 100 brand x widgets and 100 brand y widgets, you may have a defective model of brand x. Or you may have gotten lucky and gotten a good model of brand y.

The only useful conclusion that can be drawn from an experience like yours is that _that model_ of Seagate drive is less reliable than _that model_ of Hitachi drive in those specific respective environments, and under theose specific respective usage patterns. That's it.

Re:HARM

[drinkypoo]

We had a batch come in for some IBM e-servers, and a third of them died within 6 months. Absolutely disgraceful. The ones we have running Hitachi hard drives are all still going.

Your anecdotal experience runs contrary to most of the anecdotes I've read, most of which say that Seagate has good reliability (I've always found it to be so, at least, in the post-ST-506 era) and that hitachi drives are all big pieces of crap.

Proof, of course, that anecdotal information is worth every penny spent on the study that produced it.

Re:300 Terabits.

[kebes]

As others have pointed out, the hard drive manufacturers are following the proper convention, and in fact (if you look into the history), HD manufacturers have been using the "factor of 1000" convention since the very beginning (since the first magnetic platters, really).

The confusion is created because people designing memory (which is naturally layed-out in powers of 2) co-opted the SI prefixes (kilo, mega, giga, etc.) to describe sizes, but redefined them as "1024" (being a power of 2) instead. This is in complete contradiction to the well-established (and much older) SI unit conventions, where kilo, mega, etc. are always well-defined in terms of factors of 1000.

In order to cut down on the confusion, international bodies suggested that new prefixes ("kibibyte", "mebibyte" etc. ... which mean "kilo-binary" and "mega-binary" and so on) be used when one is using the binary ("1024") convention. This suggestion was ratified by, and accepted by IEC, IEEE and NIST (U.S. National Institute of Standards and Technology).

An excellent explanation, with pointers to the appropriate IEC and IEEE documents can be found on Wikipedia. Note that this convention was ratified in 1999! It's been over 7 years and people are still abusing the terminology!

As a scientist, I've always hated the confusion and ambiguity caused by using the SI prefixes to mean two different things. We have a proper convention in place, now it's time for people to use it constantly and consistently! The hard drive manufacturers are doing it the "right way"... it's time for others to follow suit. In particular, the operating system should be reporting sizes properly in "KiB" (kibibytes, 1024 bytes) or "kB" (kilobytes, 1000 bytes) consistently. I know that, for instance, Konqueror in KDE does this the right way... but I think Windows Explorer still does not.

As geeks on Slashdot we need to spread the word!... or at the very least not comit this age-old mistake.

Re:37.5TB HDD

[Agripa]

How does the capacity of a drive have anything to do with its seek time? Seek time is a function of how quickly the read arm can cross the radius of the platter, and to a smaller degree how fast the platter spins. The article claims they will be increasing storage density using this HARM thing so that more bits can be stored on the same amount of surface area. Seek time should not change significantly unless they make the platters larger, or spin the drive at lower RPM.

For a given amount of data and drives with the same physical dimensions, doubling the track density will halve the seek time excluding other factors like settling time which will become slightly worse. Track density probably goes up at about 1.4 times the areal density. Another way to look at it is that the data takes up less space on the disk so the heads do not have to move as far to reach it. None of this affects the rotational latency which adds to the access time.

Back when seek time was a majority of the access time, switching to a higher capacity drive made for significant improvements in performance. These days the seek time and the rotational latency are about equal so higher density drives are only marginally faster.

You're confused.

[Gordo_1]

When discussing the science of storage technology (densities per area and the like) researchers have always used bits. This does not mean manufacturers intend to market such drives using bits.

The willingness to confuse megabit and mebibit in order to mislead consumers is a separate phenomenon.

Expect lots of hard drive failure/read errors

[madhatter256]

Seagate!?!? Expect lot of HDD errors. I can open up every single, brand new seagate hard drive I have in stock where I work (small family owned system builder) and by using SpinRite 6.0 it will tell me the plethora of seek errors and ECC errors this hard drive is causing when you do a SMART status report. Of course, when you use Seagates diagnostic software it will tell you that the hard drive is working fine. It does work fine, for now. I went thru 2 seagate HDDs, both 7200.9 and 7200.7. Both of those, kept giving me eccerrors according to the SMART status report. The 7200.7 HDD just started to be really slow in accessing data. When it would use to take me 2 minutes to boot Windows off of it it, it takes over 5minutes, and it was the HDD that was causing the problem. Seagate is notorious for this as I am finding out. I am no longer recommending seagates, even though they have a long warranty but don't expect the HDD to even last within that warranty if you plan on using it for RAID purposes.

I'd go with Samsung, and Western Digitals. I'll even say to try out hitachi HDDs even though they are loud. Don't buy seagate unless you want an HDD that will degrade over time over regular usage.

Re:Expect lots of hard drive failure/read errors

[feld]

http://grcsucks.com/spinrite.htm

Spinrite sucks. It's crap. No wonder you think your drives are bad.

Obligatory

[Nimey]

You must be new here.

What I really need is 40 GB that is reliable

[baggins2001]

I have needs for 500GB hard drives, but I don't trust them as far as I can throw them. So I have RAID configurations for those. All well and good more data means a need for a bigger hard drive. And with all these people ripping movies they need bigger hard drives.
But what I also have is a need for a really reliable 20-40GB hard drive. On 80% of my systems I put the OS on one drive and data on everything else. So I really don't need is a cheap 100GB hard drive for the OS. What I need is a very reliable 20-40GB for the OS?
Within 3 years though I bet I am running my OS from DVD and putting my log files on flash drives.

Re:300 Terabits.

[WrecklessSandwich]

The real problem is that Windows is reporting drive and file sizes in GiB (while making the mistake of labelling it GB of course). Solution: Insert an option in Explorer (and any other file managers that do this, not sure off the top of my head what does and doesn't for other OSes) to toggle between GiB and "true" GB. Hard drive vendors should then use both units to avoid confusion. When I go on newegg or wherever, I should see drives labelled as 80 GB (74 GiB) Seagate SATA blah blah blah.

You're as stupid as Bill Gates! "640K"

[ClioCJS]

No one would ever need more than 640K memory, right? If you can't imagine why someone would use space, take your unimaginative brain and make your unimaginative comments elsewhere. I have 3.5TB of space spread across 4 computers, one computer alone has 1.5TB. Almost every drive has less than 100G of free space on it -- which is not much by my standards. Backup 4 movies in 1 night? Just used 32G. Now, my latest drive, a 750M, is only half full. Mostly with stuff that had to be moved off a 120G drive that crashed, backup copies of all my program installers (40G due to me imaging any cd-only stuff to my harddrive -- finding cds is impossible), and stuff I'd burned already but don't want to delete.

Seriously..... Next think you're going to say: "My car has 4 wheels, I can't see why anyone would need an 18-wheeler." Use your fucking imagination.

Oh really? Seagate warrantee is 5 yrs! Rest are 2!

[ClioCJS]

So you would voluntarily purchase a harddrive from a company that only has 2-year warrantee, versus one that has a 5-year warrantee?

Me, I'd rather have the 5-year warrantee. All data should be backed up. If your drive fails, buying a new one sucks. 500G * 5 years = 2.5T/yrs. 500G * 2 years = 1T/yrs. I'd rather get 2.5 times my storage-over-time. Especially after all the WD drives that crashed. (My last harddrive purchases, in reverse chrono order, by gigabytes: Seagates:750,500,500,400,300,250,200, WDs:120,120,120,120,80,80,80,60,40,25,17,4). Out of all of those, WDs have generally not lasted as long (crashed: 4g, 60g, 80g, 120g), and the warrantee has been the deciding factor of whether I need to spend ~$300 for a new drive or not.

Have you tried Knoppix?

[Vr6dub]

This is a little OT but...I have had great luck in the past recovering data from a Linux Live-CD. This is on drives that were completely unreadable in Windows, occasionally the drives are so bad they prevent Windows from booting even as a slave drive.

My experience has been, 100%, if the drive is readable in the BIOS then Knoppix has been able to mount the drive and copy data. YMMV

TB or Tb

[IceCreamGuy]

In two places it says TB, or terabyte, and then in the body it says terabit, although I suppose it really doesn't matter since the breakthrough is simply the order of magnitude, but which one is it?

Re:Reliability?

[jamstar7]

Well, if it doesn't work, just use a bigger HAMR...

Tim Taylor would be proud. :)

Screw that, just rewire it

Re:IEC 60027-2 vs SI

[Grishnakh]

Sorry, but this is bullshit. You talk about 1999 like it was a long time ago. I'm guessing you're probably 16 years old and haven't been using computers very long, but the rest of us who have been around them for longer know that we've been using powers of two since the 60s and 70s, longer than you've been alive. Just because some morons at the IEC decided to team up with some marketing assholes at hard drive companies and try to redefine the terms doesn't mean it's right. The IEC can go screw itself.

Re:Unit of measure is correct

[not-enough-info]

300 terabits = 37.5 TB (terabytes)

Did you actually have a point?

[ClioCJS]

Or was your post an equal waste of time as well?

My new blinking home system

[roman_mir]

I just got myself a new fileserver at home to play movies off of and to serve files. It's a 14 bay black tower, at this point with a 6600 dual core Intel, 4GB of RAM, 2 videocards, one is an ATI with 512MB and a capture card, an SB of some sort. I have a seperate 16 channel RAID controller on it (700CAD, is a bitch) and at this point I have 6x 500GB SATA drives mirrored. Each drive sits in a nice easy to switch on/off and easy to remove enclosure with an LCD, that reads master/slave/heat (C/F)/access indicator/health indicator and an alarm. There is an LG DVD writer there and I am transferring all of my (legal mind you) DVDs onto the beast. I use DVD Shrink and each movie in wide format without special features and only with AC3 5.1 sound and English subtitles takes about 4.4GB. I transfered about 90 movies so far, it takes between 15 and 25 minutes to transfer a movie.

The cool part of-course is that the drives are mirrored. I only have 1.5TB total space but once with a 16 channel RAID controller there is room for 10 more drives. Oh, and the beast eats 750Watts of electricity.

Re:My new blinking home system

[roman_mir]

yes, stupid me, putting together a machine to serve DVDs to my TV, using a capture card to capture shows off of cable.

Did you post that just to prove to us how stupid you are? - not only that, it also shows how poor you are.

Re:My new blinking home system

[drsmithy]

I just got myself a new fileserver at home to play movies off of and to serve files.

I certainly hope you're doing more with that machine than just serving files. I do so hate to see hardware going to waste...

On a less philosophical note, if you're using Linux, go with software RAID6 over hardware RAID1. For all practical purposes the performance and reliability will be just as good (better in some cases) and you get about a TB more space. You also have the benefit of not having to track down another one of those controller cards if it breaks.

Oh, and the beast eats 750Watts of electricity.

No it doesn't. You might have a 750W PSU in there, but I doubt it's using more than 250 (if that), even at peak load.

Re:My new blinking home system

[roman_mir]

First step is to get good hardware, I'll find plenty of applications for it. But I don't see anything wrong about using this setup as DVD server and a PVR. You are correct about the PSU of-course. I hope it doesn't actually go over 400 after I add all of the drives to the empty slots.

Re:300 Terabits.

[Grishnakh]

An excellent explanation, with pointers to the appropriate IEC and IEEE documents can be found on Wikipedia. Note that this convention was ratified in 1999! It's been over 7 years and people are still abusing the terminology!

You're another one talking about 1999 like it was a long time ago. We've been using the "1024" convention in computers since the 60s-70s. SI meanings for "kilo" and "mega" are irrelevant; computers work in binary, not in decimal, and more importantly, this is a de-facto standard that's been around since computers were invented. The only reason they're trying to change it now is because the HD marketing people want to use bigger numbers in their ads.

Here's an analogy: in the world of guns, one still very popular but very old (> 100 years) revolver cartridge is the ".38 Special", which was the standard police revolver for many years. The actual caliber (inner diameter of barrel) is .357 inches, which is why you can fire .38 Special ammo in a .357 Magnum revolver (the Magnum cartridges are slightly longer so that you don't do the reverse, firing magnum ammo from a .38 revolver which can't handle the pressure). Why then is it called .38 instead of .357? Because back in the 1800s, they used the diameter of the loaded cartridge case, not the "caliber". If we did what you and other IEC-followers suggest, we'd have to rename a lot of guns and ammunition for no good reason at all, just because some people are too stupid to understand de-facto standards and historical significance.

Really, is it so hard to understand that "kilo" is "1000x" for everything except computer jargon? We were all doing just fine with that for 30+ years until some morons decided to change it.

Yes please.

[peeg]

I need to get me one of these. Is this really feasible though? Within the time they say, that is.

Re:37.5TB HDD

[Shotgun]

If I redefine a mile to be 3,000ft, I would be able to run twice as fast as I do now.

Cut the bit size in half, you double the number of bits that can be crossed in a given time.

MiB

[AJWM]

I thought MiB was a secret organization dedicated to keeping tabs on extraterrestrials living on Earth?

Is GiB their informal name (guys in black), or what the female members (gals in black) go by?

Re:decrease platter size

[Grishnakh]

They're already going to 2.5" SAS drives for servers. I have a bunch of them in my lab here; they're usually 37GB per drive, and spin at 10k or 15k rpm.

The problem with smaller drives is much less area on each platter, and the linear speed at the outer edges of the platter is much less. However, this isn't that important because they use perpendicular recording, meaning the data is more densely packed on the disk. Also, the big advantage is significantly lower power consumption, something that's much more important these days. It takes a lot less power to spin so much less mass.

In addition, with RAID now a necessity on servers, the maximum speed of each individual drive isn't quite so important (since they're working in parallel), and the combined power consumption of a large array (16, 32 drives or more) is very important.

Relative durability

[Beryllium Sphere(tm)]

Is that because disks are usually in service all the time? Has anyone seen numbers on the failure rate of disks sitting still in climate-controlled storage? There have been reports that the spindle lubrication gets sticky if left to sit for too long. Is that fact or rumor?

SCSI drive

[WMD_88]

Sounds like a SCSI drive would suit you. You'd need to buy the controller card, and the drives are expensive, but if you want a reliable, smaller drive, that's it.

Re:300 Terabits.

[Grishnakh]

Sorry, but you can't just go changing standards willy-nilly and expect people to accept them. 1999 wasn't that long ago.

Re:Reliability?

[ksd1337]

Stop! HAMR time!

Re:Why does everyone mention defrag times?

[WMD_88]

Now we have NTFS which doesn't require defragmenting.

Since when? Nearly every Windows machine I work on (owned by laypeople) with NTFS has a badly fragmented hard drive. I can't say if the drives are faster afterwards (I don't run benchmarks), but they are still pretty fragmented.

Re:300 Terabits.

[toddestan]

As others have pointed out, the hard drive manufacturers are following the proper convention, and in fact (if you look into the history), HD manufacturers have been using the "factor of 1000" convention since the very beginning (since the first magnetic platters, really).

I still have a 20MB hard drive that holds 20,9xx,xxx bytes on it. The switchover happened back in the 80's, and was a deliberate move by the harddrive manufacturers to deceive people. You can rattle on about standards all you want, but it all started because of a scummy marketing move.

Besides, they are still only one of the few playing that game. When was the last time you saw either a 528MB or a 512MiB memory module for sale?

Re:"Within a matter of years", but still significa

[francium de neobie]

I don't think it's too far-fetched. Indeed it fits quite nicely in the past trend of hard disk size growth.

My numbers aren't exact, but they shouldn't be too far from truth:
I remember in year 1995 I was looking at 850MB-1GB hard disks being mainstream.
Then, in year 2001 I was already looking at 100GB hard disks at the high end, and 40GB being mainstream.

So it took only 6 years for 850MB to become 40GB in the mainstream market.
Now we're already having 750GB at the high end, I would be very surprised if we still can't buy a 37.5TB hard disk a decade later.

Re:I Cannot Believe

[thedeviluknow]

Umm dude? M = Mega (1,000,000) K = Kilo (1000) :)

Re:IEC 60027-2 vs SI

[dangitman]

Sorry, but this is bullshit. You talk about 1999 like it was a long time ago.

It wasn't. But the amount of time the "kilo" and "mega" prefixes have been used makes the time that computers have existed seem like the twinkling of an eye. It's the computer people who are the newbies here. The "base two" shit doesn't even make any sense. Just because you are counting in base two, doesn't make 1024 a "kilo." Use a fucking different term to refer to 1024, rather than using something which has always meant 1000.

Re:IEC 60027-2 vs SI

[totally bogus dude]

I'm guessing you're probably 16 years old and haven't been using computers very long, but the rest of us who have been around them for longer know that we've been using powers of two since the 60s and 70s, longer than you've been alive.

I'm guessing you're probably 60 years old and haven't been using metric very long, but the rest of us know that the metric system (from which the "kilo", "mega", "giga", etc prefixes originated) was in use well before the 1960s. Computer scientists should have known better than to overload the meaning of existing prefixes, and that "close enough" isn't good enough.

A quick search of the internets tells us that kilo was "Officially adopted in 1795 (though in common use before that), it comes from the Greek "khilioi", meaning thousand.". What kind of "scientist" takes a word that has for hundreds of years meant 1,000 and decided to make it mean 1,024?

Most of the others are listed on Wikipidia as being "confirmed" in 1960, which suggests they were in common use well before then. Even if not, the comp scientists of the era (you, perhaps?) should have at this point seen the problem and stopped overloading the meanings of these terms.

Regardless, saying it shouldn't be changed because it's "always been done like that" (even if "always" is only 50 years) is not a good reason. We have a good solid history if completely ignoring security when designing computer systems, especially for the mass-market. That doesn't mean it wasn't a mistake, and it doesn't mean we shouldn't try to rectify it.

Computer technology is still in its infacy, and we should seize the opportunity to fix things before they become irretractably entrenched in everyone's everyday lives.

Disclaimer: I like "1 kilobyte = 1,024 bytes" (partly because it makes me feel clever, like a real computer scientiest working in base 2 and everything!), and I don't like the KiB notation or term "kibibytes" (because it sounds stupid), but I dislike ambiguous units of measurement even more. And make no mistake: "KB" is an ambiguous unit of measurement.

And more reliable

[WindShadow]

It would seem the new technology would be more reliable (when the technology matures a bit) since the new magnetic media appear to have better hysteresis characteristics. Once written correctly the data should stay written, so that in the extreme case the platters might be recovered even if the mechanism or electronics failed. Obviously only practical when the data have significant value.

The bad news?

[StikyPad]

Transfer rates will still hover around 60MB/s, and running chkdsk on your 300TB volume? 2.75 years, on average.