Slashdot Mirror
Seagate Spins 15k RPM HDs
An anonymous reader sent us a story about Seagate spinning 15,000-rpm disk drive. This stuff spins faster then my head ;) I don't shop for hard drives very often... it kinda blew me away to see 40 gig IDE drives for only a few hundred bucks. I'm getting all nostalgic for the days of two 360k floppy drives. Weird.
-=-=-=-=-=-=-
This signature contains text from the worlds funniest signature.
But how fast do the HEADS move? That's the other most important side of the equation...
If you could be told what you can see or read, then it follows that you could be told what to say or think - BoC
I thought drives faster than this already existed.
Seriously though. That's going to kick ass. Now if we could just kick up the data transfer rate faster than Ultra66 (and make it easier to use) we'd be set.
How loud is the thing? The 7200RPM drives usually whine a lot...
kwsNI
I remember the days when floppies were the size of records! Those were the days!
That's a weird way to spell weird....
Why, I can remember the days when I used a wax-based 33rpm storage system! None of this 15,000 kind of stuff for me, nosir, it's LPs or the highway.
Just how fast can current technology take us with hard drives? What are the other technologies being developped to cope with speed limitations that we'll certainly hear about in the near future, not to mention to lower the price of the current hard drives out there?
EraseMe
I've done some reading about these new drives and word is that the super high RPMs generate so much heat that Seagate has taken to routing coolant through the drive heads just like blood through your body. Of course the little coolant pipes are tiny tiny, but they're there, chillin out, to use the parlance of our times.
Wow, that's really fast...the engine in my car would blow up if it even thought about going that fast.
What is the access time on a Hard Disk like this? If it's not much faster than the current 7200 drives, i'll just hold out
-Tim
You know what pisses me off most about this stuff? Every time I shop for hds for my Linux box (I don't do IDE, except for spare storage space), they require a whole new SCSI card.
Back many years ago, I started with a simple $15 NCR SCSI2 card and two or three 2g Seagate drives. That was great for a while, then I wanted more. After shopping around for quite a while, pretty much anything over 4g was Ultra-Wide. Ok, so I plunk down the money for an UW card and a couple 4.5g IBM drives. Now, I figure it's time to upgrade again, and guess what? Most all the 9g drives I see are Ultra2 Wide. I'm going to have to talk with Al Gore to stop inventing these new technologies so dang much!!
Jesus, this is fucking comedy! You can't pay to see this. It's priceless. :) Classic. This needs to get laminated and hung above the urinal in the Holland Mich public library mens room.
And I used it in the early 1970s. Univac SS90 (90 column cards, round holes) had a 50,000 digit (5000 words x 10 digits) 17,000 rpm drum for main memory. Now Seagate is getting close, but they aren't up to that yet, and by gum, never mind that Seagate claims all the other manufacturers are 2 quarters behind: they are all 42+ years behind Univac.
Dang thing took an hour to spin down and ten minutes to spin up.
--
Infuriate left and right
I remember my old BBC Micro. A single tape drive that was pretty unreliable at 300baud.
Mind you, it had a BASIC interpreter and a neat 6502 assembler in 16K of ROM, and would boot in less than two seconds.
They don't make them like they used to. Fortunately.
(Incidentally, I've still got it. Must see if I can get it to run NetBSD some time.)
Now that's fast. I remember when 60ms was fast.
Now we're talking about a mere 18G drive, but I suspect that such speeds will probably be available on larger drives shortly. (I don't have any inside info on that, though.)
Men there sit down to urinate. Showing your penis as you urinate is exhibitionism, which is punished by public flogging in Holland, Mich.
Why even go that far? How about he tries a little COMMON SENSE and PROOFREADING? The shit is plastered all over the pages "use preview! Check those links" I don't think cnn.com is racing to post some geeky crap like this on their main page.
Progress in storage follow Moore's Law rather closely for both size metrics (Areal density)
and performance (data rate). It is easy to over state capacity of a drive technology as the manufacture can just add more disks to up the drives capacity. The real magic is increasing the areal density of the recording media. The cost of building a drive is basically fixed. Heads cost X, disks cost Y, etc. The inductry adopts the next generation technology when it becomes cost effective.
The business model of the drive business is crazy. It take 18 months to develop a drive and it has a market life of just 6 months. The manufacturers FLY the drives to the US using cargo 747s. Also the profit margins in the drive business are razor thin.
The drive have also evolved some very cool tech over the years. If you kill power on a drive the motor becomes a generater and powers the head into the landing zone. Today's drive include either a ARM7 or 16 bit DSP class processor. As long as you don't shock the drive (1/2 ich can kill a drive) it will last forever, unlike drives of old.
Scott
Thank you for bringing some charm into my dull monkeyless life.
Bravo, good man/woman/beast.
Yeah, yeah, we've heard it all before. Magnetic storage has been around way too long. Break out the 4000gig holographic cubes!
The interfaces (SCSI, IDE, Firewire) have been able to handle uncompressed broadcast quality video for a long time now... Until just recently we've needed huge disk arrays in order to handle those streams. This will probably be the first drive that can handle that on it's own (27+ MB/Sec).
Of course it's much easier to insert faster crystals into PCI cards than it is to make the jump from 7200 to 10,000 and now 15,000 RPM's... Sad thing is, i think all but one of my drives are 5400 RPMs... But next time i need one, i'll probably just buy on the high end again.
The big question is how loud is this thing? I mean, the 10k drives i've experienced are pretty loud - loud when spinning and it sounds like someone knocking on the door when it's seeking... I can't imagine something even faster being any quieter.
--onyx
--onyx--
Why isn't this called a 50X drive? That's how much faster it is than a 5.25" drive, and that seems to be how other drive speeds are measured, based on the speed of the first popular drive of the type... :-)
A 3 inch platter rotating at the speed of light would be doing 3.183e7 rpms. So, no faster than that.
And damn does this thing fly!
Everybody who drools over cool new tech are going to love the storage options coming out in the next few years!
Later.
Enforce Darwinism
Crap, that stupid
I can see it now. They're going to get our hopes up like this, we're all going to buy, buy, buy, while they sit, greedy eyes staring into nothingness... "Sure, it's 15k RPM, but we never said it was ultra-ata 66" (evil laughter) Them major corporations are tricky. -herd
Ha! Seagate beat you to it. You should have worked more instead of sucking each other's dick.
right after Roblimo ate your balls....
I hate to be around one of those things without ear protection. Earlier this week, an old, slow, Barracuda crashed its heads on me, and I dropped in a 7.2K or 10K Hawk (forget which one) as a replacement. Man, that thing sounds like a jet engine when spinning up.
The story doesn't mention the Db of the 15K drive. The noise from high speed drives almost precludes them from being used on the desktop, if you keep your machine constantly on. Now that I need to get another drive, for a "semi-hot" standby replacement, I'll definitely be asking the ambient noise level before I get anything.
It's nice to see Seagate getting back into shape. Back in the ancient times (mid-80s), they were the undisputed king of hard drives. They almost bit the dust, but are now making some pretty good iron.
--
Enforce Darwinism
Crap, that stupid
We all know that speed isn't everything. We want large drives that are cheap. But also smaller drives that are fast. Becasue for most people the amount of data that needs to be accessed quicky is small (eg os) while the other data (eg all those pirated games and MP3s) dont need the speed. We want a cheap 100gig 3600 rpm drive + a damn cast 15k 4.3 gig. Keep us all happy
I wonder how long "trickle down technology" will take to get this to John Q Public? It is unfortunate that technology like this is never released on a broad basis. How long do we have to wait for someone to make something like this in a large scale? Whoever does so will quickly capture a substantial portion of market share if they can just make enough...
One thing that has interested me for a while is what is the chance of a HDD fragmenting. I dont mean no data defragmentation. I mean physical defragementation.
At 15,000 RPM can a HDD case contains the pieces. Even if the chance was 1 in a million I would like to know if I should put some more steel casing around my drive bays.
A friend of mine has lost feeling in his foot from a flywheel in his RX7 breaking up, it broke through the bell housing, through the steel floor then through the carpet breaking his tibia. I know it is different but a piece of HDD platter could do some serious damage.
--- Can i borrow your Clue-Stick(tm)? I need to go beat a few people with it...
If the sounds of these drives are anything like their Cheetahs we may need to sue them for tinnitus aggravation.
>Try reading the article instead of trying to get a fast post (#6 in your case)
:-)
:-)
Why read when I can just press (in netscape 4.x): Ctrl-F, "head",
and find nothing (but ahead... bad pun, sorry)
BTW: I did read it... there isn't too much to read. If you can't read that in 5 minutes, might I reccomend a speed reading course?
Sure, they say 3.9 ms access, and 2.9 ms, and 2.0 ms, that doesn't rule out a head that takes 1/10th second to travel from extremity to extremity... Maybe spindle rotation is the only major important thing in storing uncompressed live motion video, but what about fileserving - for that I don't much care about RPM, I care about full head seek time.
Noise would be important for me too - I like to listen to music while using my computer. Noisy hardrives make that difficult.
If you could be told what you can see or read, then it follows that you could be told what to say or think - BoC
Now that's fast. I remember when 60ms was fast.
Several years ago when I worked for the university, I helped throw out an old word processing system that my boss insisted was outdated (it wasn't broke, so why fix it?) It sported an old 10MB hard drive and if I remember right it was powered by a three phase motor. I laugh when a person today says installing a hard drive is complicated. Today's drive weighs less than 100 pounds and doesn't require a special circuit breaker.
Makes me want to install my advanced MFM card and see how well those state of the art IBM drives will work with my 2.2.12 kernel. Does anyone still know what RLL means anymore?
...as interesting instead of funny.
:-(
(Teach me to forget and bracket a word with the less than WORD greater than symbols. Last I checked there wasnt a <funny> tab in the HTML specification.
Sorry...
It actualy exist, or at least have existed. I remember that a friend of mine had a LP where the last track was a program for a Sinclair Spectrum. The program was ment to be started at the same time as you played the music and wow, you get a music video. (The Spectrum normaly stored programs on an audio casette recorder, through a standard line out/line in.) I don't remember the title on the LP, not even the artist/group.
If you can't stand the risk of getting killed by your hard disk, go and play with dolls instead.
they are???? I learned to program FORTRAN on a sperry univac that had a DRUM as a harddrive....
lol
RLL==RUN-LENGTH LIMITED
Source: the Maxtor Hard Disk Glossary
...Open Source isn't the only answer -- but it's almost always a better value than the alternatives...
it was so on topic. you're being completely insensitive to the victims killed and maimed by out of control hard drive platters, which can gain intense velocity as they rip through the metal casing of the hard drive and slice through anything in their path. unstoppable juggernauts of destructive whirling power that can cut through flesh and bone like butter and become embedded in walls, still slick with the gore any unfortunates in their path.
you cruel monster.
The speed is nice but I read an article on AnandTech.com that says that to acheive such speed, they had to reduce the plater size considerably. Meaning that these drive are curently limited to 18 gig. Not bad but it still makes you lon for those 40+ gig monstrosities that are comming out.
For a preview on the drive at (duh) www.storagereview.com
>That's Alt-F. Oh, wait. You were using Windows, right?
:-)
Sure was...
(Not that I don't like Linux, I use it all the time, but I decided to watch a DVD while I posted that...
If you could be told what you can see or read, then it follows that you could be told what to say or think - BoC
HDD platters have abou the same stiffness as a pice of cardboard you lamebrained shithead.
At least some interesting comments.
wwwwwwwwwhhhhhhhhiiii iIIIIIRRRRRRRRR!!!!!!!! (rumble*rumble*rumble*BrRrRrRrRrRrRrRr...)
And the pc case (before I glued the rubber feet back on) would actually start moving towad the edge of the desk!
If the drive heads warmed up less than the platters, the differential expansion due to thermal changes would surely distort the spacing and change the character of the way the heads ride over the platter airflow. A difference in the temperature between the air and heads could also be a source of potential problems. I have doubts they specifically cool the heads. But perhaps they do have coolant running through everything, or maybe the outer frame.
The heat sources would be the electronics (mostly underneath, but some are inside, such as the head pre-amps), the platter motor, and the voice coil. The better the bearings are, the lower the resistance to spin, and the less energy required to maintain RPM. But at higher RPM, the resistance increases by some formula I have long forgotten, so there will still me more energy needed, and thus more heat dissipated, to maintain RPM. Lighter platters would also help, but I'm not sure just to what degree this is once the drive has spun up. Head seeking needs to be faster and faster to meet our demands and expectations, too, and that means more energy in the voice coil to increase the acceleration.
So, they will be very hot! But will the heads specifically need to be cooled? I doubt it. And running coolant out to the heads would likely weight them down a whole lot.
now we need to go OSS in diesel cars
Heck, Honda hit 17,000 on their NSR500 a while ago. So, why can't hard drives spin faster than a 200-lb motorcycle engine?
Just food for thought. What's the design bottleneck?
Carefree highway, let me slip away on you.
When I think about getting a 10,000 (or now 15K) drive for home use, I wonder about the additional cost in electricity usage if the drive is left on all the time for each year of ownership. Does anyone have the information to calculate this or a good feel for what an estimate might be?
You'd think the advantages of drums would still apply today. I bet the data density and sustained read would be a lot better (and obviously more consistent) in both cases.
Are discs just that much cheaper or smaller, or what? I mean, a drum wouldn't fit nicely in the drive slot, but they might come in handy for high-performance web servers.
I bet if you had some nice solid drum drives running at that speed, you could mount them in your car and use them as flywheels for regenerative braking and to hold the world's greatest portable mp3 collection.
MFM still works fine. I was running a 486 with 2 40 MB MFM drives (One IBM, one seagate) a short while ago with 2.2.0(pre8). One was /home, the other for / (with boot, root, bin, sbin, var, etc), and the rest was, shall we say, NFStory. NFS over arcnet [the network attached to this computer] was faster than these crappy things, anyways.
FYI: This was with a Full length WD MFM card.
Fun - I remember the cool sound they make. And I still have them. And the very best - a broken one with a serious head problem - it won't stop violently moving it's heads back and forth. A 5 1/4" full height drive doing this means it will move off your desk in 2 or 3 minutes.
Nothing scares people more than when I say "hold this for a minunte" just before I power on that drive. Haha. Almost as fun as throwing charged capacitors at people (I must have been _real_ popular, huh? >:-)
If you could be told what you can see or read, then it follows that you could be told what to say or think - BoC
If your car's engine was the size of a chainsaw engine, it could go that fast and more. Its all about the forces of rotational mass and moving more bits per second through smaller spaces. If it can be made in a smaller space, higher speeds are possible, and it can do more of what it does best each revolution.
A pair of massive turbochargers on a V16 deisel CAT generator has a maximum rated speed of 60,000 rpm. A pair of tiny turbos under the hood of an RX7 can routinely see 150,000 rpm.
The calculation of the rotational speed that gives the maximum access speed is left as an exercise for the reader.
Given the utterly clueless replies in this thread, I suppose I should clarify that I mean the advantages of a drum compared to a disc using the same modern material science advances etc. NOT the ancient main memory drum vs. modern RAM or even against a modern HD.
I have a Seagate 10,000rpm disk that pulls 27mB/sec easily on the bonnie test. And that's with a tagged queue depth of only 8 on a UW controller. The drive supports LVD and much deeper queues, so I'm sure it can go even faster.
Sure, faster disks are going to be nice. But they're going to be pretty demanding on the bus. If a 15,000rpm disk is even 50% faster than a 10,000rpm disk, there's not room for many neighbors on a LVD bus. Forget UW SCSI - it'd be a bottleneck with even one of these disks. Looks like it's finally time to break out the fiber...
Then we'll _never_ get to see Internet 3!
Sure, I have a thankless job. That's okay. I have a lot of (non
Also, does anyone know if they have any IDE controllers that can RAID 1 two identical drives together without using a software driver to do it? Most of them I've run across implement the redunancy in software, which I just don't trust, no matter what OS it happens to be.
Now, if they could just RAID 5 some IDEs together on a card...*homergurgle*
Wow, 15,000 RPM's. Putting that through my trusty calculator, thats 250 R per second. Pretty soon these things are going to be going so fast they'll injure someone if they get loose and fly out. Just think about the damage one of those things could cause inside your computer. :)
-- Moondog
You can watch the bits come off the platters.
Oh, and it's "Weird".
Back when I used to ride, a typical sport bike engine lasted 40,000 miles, which is perhaps 600 hours of operation. A hard disk has to last 10,000 hours.
You will give us a valid credit card number. You will pay us $N per month. If you do not pay, we will use the stereo microphone as a proximity sensor and eject C: right into your cheap-ass pirating gonads.
DO NOT open that link if you could get in trouble (e.g. FIRED!) for viewing p0rn over the corporate network. (Like I can.)
If you voted for Nader, THIS IS ALL YOUR FAULT!!
(This is getting silly, but hey, why not? It's only karma).
All facts from the article:
>average seek time of 3.9 milliseconds
Translation: Bullsh*t seek time. Doesn't mean nothing. These numbers are always faked.
>data transfer rates up to 48 Mbytes/second
>15,000 rpm
these don't count...
> A 15-k spindle speed helps bring latency down to 2 milliseconds on the Cheetah X15, he explained, compared to 2.99 ms on the company's 10-k rpm class drives and 4.17 ms on its 7,200 rpm drives.
ie. on the same track. Big whoop, still nothing about head speed.
>And a new, speedier actuator design brings seek time down to 3.9 ms, compared to 5.2 ms for the company's 10-k drives.
Kinda sorta counts. But still doesn't tell me anything about how fast the heads are (but they are just faster, I'll guess I'll just have to trust them. Right... 8-P )
>The 15-k drive benchmarks at 140 I/Os per second, Hood said, using Intel's IOmeter benchmark, compared to 84 for 7,500-rpm drives and 105 for 10-k drives. All in all, by Hood's reckoning, the 15-k drive delivers a 33 percent increase in I/Os per second over 10-k drives and a 28 percent improvement in time-to-data.
[sarcasm] And I bet it scores high in MIPS and FIPS too... [/sarcasm]
Now how fast are the heads? They barely get a mention in therem, and the mention they get is useless, and I just PROVED I read the WHOLE article.
It seems all anyone cares about is rotational speed. That's great if I want to rip DVDs to AVI (not that I do), but it isn't a big deal for fileservers (within reason - 3600 RPM obviously is going to be a major performance hit). And most of us use our HDDs for random file access, not watching big huge uncompressed movies.
I want to know how long it takes the head to go from track one to the last track. Things like that would be nice, not vaguely defined things like "average seek time".
If you could be told what you can see or read, then it follows that you could be told what to say or think - BoC
Comeon man,
I've got a stock '89 Volvo 240.
The engine has never been rebuilt, and pretty much the only work done on it is the standard every 10k maintence.
My computer is worth more than it is (but the car does a good job carrying my computer to lan parties). I don't think that i'll be putting a turbo in it anytime soon.
-Tim
Yes. This is a tank.
If you're gonna open up a hard drive, you might as well take out the NIB magnets. NIB magnets are the most powerful permanent magnets known, and they are used in hard drives. You can do all kinds of neat things with them. They are so powerful, that you can build a compass just by setting one of these magnets on smooth surface, such as a plate. They will overcome the surface friction and point north-south! Also try dropping a NIB magnet between 2 closely spaced big aluminum CPU heat sinks. They induced eddy currents will cause the magnet to descend rather slowly.
Have fun!
Ever look through your browser's cache directory? You'll swear there's stuff in there you've never seen before.
I've found it difficult to find objective reviews on storage issues. The best resource that I have found (and perhaps the only one) is
http://www.storagereview.com/
A fitting url =]
They also have a small blurb about the new seagate drive, they are always on top of the latest innovations in hard drive tech.
Before you buy your next HD... read their reviews... they'll never steer you wrong.
-ecc
15,000 RPM with an ~3" drive = 4500 Gs (g ~= 9.8 m/s^2 for those who failed physics 101). Wowzie. That sounds painful. Does anyone know what the material limits are?
'pologies for AC - my account seems to be trashed.
Am I the only one that would be curious to see a table of hard drive sizes, speeds and costs plotted from the 1970's through to now & projected forward for the next five years ?
I would appreciate the URL(s) of any existing tables/data, failing that perhaps slashdotters could post their own recollections so we can make up a table of our own.
I'm currently coding in the area of GIS / composite aerial / satellite imagery & terrabyte systems are starting to look pretty common place here - how long before we see the first low cost home terrabyte compact drive ??
Just curious.
:)
LW
What we need is one head per track!!!!
I want my disks to be able to handle 200MB/s each.
Disks are so slow it's depressing.
Deleted
Solid state heads, one per track, stationary.
No seek time, blast the data out in parallel.
Deleted
This puppy is fast, slick, quiet, and cheap. Not cheap like IDE cheap, but cheap for SCSI. I'm going to go buy another one:
(scsi0:0:0:0) Synchronous at 40.0 Mbyte/sec, offset 8. Vendor: IBM Model: DDRS-39130W Rev: S97B Type: Direct-Access ANSI SCSI revision: 02
Another oldie but goodie:
dbug g=c800:5
Gawd, the nightmares....
st225, 20 MB. That drive could walk, tho. I bet this new one will RUN across the table.
St252ax, so big you had to create 2 partitions, because of the 32MB partition limit imposed by, (you guessed it), Microsoft.
I'd hate to buy the new drive, just to see how well it'll walk, but benchmarks such as this are IMPORTANT!
Paul
A)ttack! F)lame! Y)eah_Flame! Q)uick_Flame_Back!
or cough m)oderate, but you dont have to you could not if you didnt wanna now get in the car and forget what you didn't just see that sheep do.
GET IN THE F^&%ING CAR. NOW !!
Hey, with these things going so fast, they can double as gyroscopes in airplanes and stuff. Well, at least for now, they can act as stabilizers for our computers.
At least not in the one I programmed. Nothing but drum. No core, no RAM. One gate per circuit card I think. Worked on it for about 3 years. Maybe they snuck it in by the time you got to it :-)
--
Infuriate left and right
200 words per band sounds familiar. No parallel reads; all serial. 17 microsecond word time.
I loved optimizing the instructions (one + one: 2 digit op code, 4 digit data address, 4 digit jump address (every instruction jumped)). You optimized it so as each instruction finished, the next instruction was coming up under the heads, and the data was right there too. Not nearly as bad as it sounds, because you had bands to choose from. Sort of like a 25 way cache. For example, instruction at 205 referenced data at 410 and next instruction at 615. Total 10 words, 170 microseconds. If you know "Mel the Real Programmer", that was a drum machine.
--
Infuriate left and right
Nan desu ka? Ja, moving fast enough, a wet noodle could be put through your skull, tovarisch. also comrade, you mispelled the word, "pice" nyet? should be "piece", da?
MTv is devil, ja?
MTv wa totemo 'gay' desu.
Sooooo offtopic. Moderate this dooooooown. TO THE BOTTOM OF THE SLASHDOT ABYSSS!! MUAHAHAHAHA!! AND NOW YOU, My Friend, Why Are You Still READING?! *SNORT* ROOAAR!
yes, you see it coming but you read it anyway:
"Why are you still....................
Seagate's spin relies on latency (2 ms) and seek time (1.9 ms) and as usual, they don't tell us how they come up with those figures.
Even if those figures are correct, that still doesn't explain how Seagate's new 15-K rpm drives would be better.
Imagine now, you have a server, it's transaction-busy, and you need to have lots and lots of io.
Would you rely on ONE 15-K rpm hd, or would you rather have 2 or more slower-spinning drives, maybe several, connected to raid-5 array, so to spread out the io load?
Think of it, willya?
A drive that spins 15-K rpm spins twice as fast as a drive that spins at 7400 rpm, that means, a 15-K rpm drives will NEVER last as long as the other one which spins half as fast.
I would rather have a full array of 7400 rpm drives in raid5 configuration than rely on the faster spinning drives that may crash before its time, and that will certainly give me lots of headache. For crashed drives means lost data, and if my server is transaction-busy, lost data means lost income.
Furrthermore, the recent MTBF from all HD manufacturers are almost always bogus anyway. How I long for the old time where MTBF means just that, Mean Time Between Failures.
Muchas Gracias, Señor Edward Snowden !
I'd stay the hell away from these new 15K drives though until at least a couple of generations from now.
-A.P.
--
"One World, one Web, one Program" - Microsoft promotional ad
"Remember when the U.S. had a drug problem, and then we declared a War On Drugs, and now you can't buy drugs anymore?"
I love the part about bringing Internet users faster searches, quicker downloads and a more enjoyable online experience. you know just the other day I was thinking this 2.88 modem connection is great but I just wish my 10k rpm hard drive didn't slow downloads so much. Please can't a company try introducing a new product with out the blatant Internet marketing plug. What do they think that every one has a 40gb fiber optic line running into their house?
They recently went out of the SCSI business and this probally hurt them. Now Seagate comes out with 15000 RPM hdds. This will hurt WDC among people that want/require the best performance because they don't have the SCSI performance any more and their IDE drives are currently very lagging compared to Seagate.
Don't call my crazy, that's what they called me back in the home!
Try one of those old bargain-bin CDs of Microsoft Golf...
makes a tremendously loud noise
I have seen the future, and it is inconvenient.
No your "2.88" modem won't overkill this drive but...
Slap these new drives in, say, yahoo.com, where the servers must slash through the database millions of times per day, and there you have it: faster searches.
Slap these drives in ftp.cdrom.com, which consistently has 5000 connected users downloading crap they probably don't need, and there you have it: faster downloads.
But seriously, this makes me nostalgic for the 1980s?
Does anyone remember which 1980s word processing spellchecker was the one that employed the "'I' before 'E' except after 'C'" rule?
Every time I wrote "weird" it would suggest "wierd."
To be fair, it was kind of innovative; instead of relying on a library containing every common English word and matching the document's words to the library's words, the spellchecker relied on an algorithm based on spelling rules. This saved a lot of space.
Only problem is that the "'I' before 'E' except after 'C'" rule is refuted by science. (chuckle)
sig semper tyrannis!
And the game you're playing is "Pin the tail on the cyber-donkey" where you post blindfolded to see whether or not you can hit the right story?
I see even classic Slashdot is now pretty much unusable on dial up anymore.
"And a new, speedier actuator design brings seek time down to 3.9 ms, compared to 5.2 ms for the company's 10-k drives." Those actuators are what move the heads around the platter. Also, this disk has twice as many heads as current 18 GB drives, which also contributes to reduced seek time.
HD Platters are quite stiff - 1/16 to 1/8 inch thick alum. That is why the are HARD Disks.
Not true. The half life decay time of certain unstable subatomic particles (muons, mesons, etc.) are well known. But when these particles are accelerated to near c speeds, their average decay time can increase severalfold. The particles appear, to us, to live statistically longer than if they were stationary. This is not speculation. It has been observed time and time again.
Hey, as great as this may sound, our quest for speed is creating hardware dangers. At 15,000 RPM, on a platter about 3" in diameter you're getting speeds around 140 mph. For those of you in metric (as we all should be) that's about 8 cm diameter, 60 feet per second, or 230 kph. That's really fast. That could break through casings and possibly sink an inch or possibly much more into soft tissue. In the right places, that can kill. Sure, the odds are low, but who wants to take that chance? (Remember the guy whose story was recently posted here because he was misidentified in a DNA match?) If it's in mass distribution, it could happen. A notable feature of Moore's law is that as technology advances, it also changes. A modern chip (Athlon, PIII, UltraSPARC, G4, Alpha, or your favorite miscellaneous supercomputer) is far more complex than the 8086 that's rusting in the next room. They did a lot more than just make it more compact for the transistors and increase the clock speed. Periodically, the technology must change qualitatively to enable further quantitative improvements. From what I can see, we should start examining more optical solutions, since that seems to be much greater opportunity in density, although it is not yet being fully exploited. Maybe holographic cubes are the answer. I recently saw some specs for an experimental quantum system (binary) that uses lasers to access data stored by bacteriorhodopsin, a chemical similar to what activates the light sensors in our eyes. They think they can make a system that would run at about PC33 speeds, with a cubic dimension of a centimeter, with 4096 storage nodes in each dimension. That's 4096x4096x4096 bits, or 8 GB, at nearly RAM speed. A few modifications, and I'll be a happy customer.
WARNING: there is a trojan on your
Does "15000rpm" strike anyone else as odd, considering that mainstream drives have spindle speeds that are multiples of 1800rpm (e.g. 5400, 7200)?
I have a feeling that this new drive actually has a 14400rpm spindle speed -- especially since the article mentions "7,500-rpm drives". Is there actually such a thing as a 7500rpm drive, or is the author of the article just plain clueless?
If the new drives are in fact 14400rpm, and not 15000rpm, and Seagate has the gall to market them as 15000rpm, then we might actually have a nice juicy class action lawsuit on our hands. (But then again, look at the so-called "56kbps" modems...)
begin 644
Get a Kenwood True-X drive. Doesn't spin up like a turbo (it's quiet!) and it hauls butt. I picked up a Kenwood True-X 42x for $50 at Fry's. That's only about $10 over a "el cheapo" 40x max drive. Killer deal.
IMZombie (Too Lazy to Log in)
Seagate lists the A-weighted idle sound power level of the 15K RPM drive (in the specs on their website) at 3.9 bel. For comparison, their Cheetah 18XL, a 10K RPM 18GB, is 3.8 bel. I have a pair of older IBM 7200 RPM UW SCSI drives, the UltraStar 9LP model, which are rated at 4.4 bel idle, and the noise level from them isn't all that bad. Noisier than I'd like but not unbearable.
rodent...
rodent...
Tactical nuclear weapons are a viable alternative!
Can't something like this be dangerous? I mean what if the disk somehow flies apart due to the extreme inertial forces from the centrifugal forces that it is experiencing. Ok, I could be just thinking to hard here but there has to be a limit somewhere...
Nathaniel P. Wilkerson
NPS Internet Solutions, LLC
www.npsis.com
Nathaniel P. Wilkerson
www.haidacarver.com
> I have been pondering the idea of having more
> than one set of heads on a HD it would double
> the transferrate and do wonders to the avg.
> seek time.
Actually, in a multiple-plate drive, there are more than one set of heads.
But that wouldn't make seek time faster though. For the HD drive mechanism is such that the heads are connected to the arms, which are connected to the servo, which is located in the center, while the plate rotates around it.
The seek time depends on how fast the servo can move the arms, with the head attached at the end, from one sector of the plate to another.
And it is impossible for one hd to have more than one servo mechanism, for any circle, there will only be one center, and the servo located in the center.
Cost wise, the heads don't cost that much either. Yes, relatively speaking, it's more expensive than the plate media, but still it wouldn't double the cost of the drive if there is more than one head for the hd.
> I understand that the heads represent a
> sizable portion of the cost of a HD so maybe
> thats why we haven't seen any such creatures.
> As dual-head drives would cost aproximatly
> twice as much as a regular drive but two
> regular dirves would also double the space
> when RAIDed.
So, currently the only way to speed up a hd is to speed up the seek time, and the simplest way to do that is to rotate the disks faster. That's why Seagate is coming out with the 15K rpm drive.
Muchas Gracias, Señor Edward Snowden !
I remember my friend showing me a SCSI hard drive a few years back they had somehow gotten to go spin faster than it was supposed to, and generate enough centripetal force to knock over the entire computer case. Quite amusing.
I wonder how long it will be until sound isn't
really a problem. They may be loud, but you
won't be able to hear them.
Indy cars (and motorcycles) have been reaching
higher and higher RPM's. There is talk that if
they go higher they may be ultrasonic.
Maybe you'll hear them as they spin up and then
the sound will become inaudible...
Has anyone ever worked out the bandwidth of one of these planes if you loaded up the drives with data before shipping. I guess the drives weigh 100g each, and a cargo 747 must be able to carry 100 tons, so thats approx 1 million drives (reality check, this is a pile 10m x 15m x 2m, should fit in easily), so a single flight carries about 18 PB (peta bytes, 10^15 bytes) of data. Say it can make one delivery every other day, that is about every 180000 seconds, so we get a bandwidth of 100GB/s, or 800 Gb/s -- who needs project Oxygen (320 Gb/s transocean cabling) anyway?
"Latency" do I hear someone ask? "Don't be small minded!" I say. Latency can be dealt with by proper caching strategies at a higher level of the protocol stack.
Steve Linton
PS a 100 000 ton cargo ship full of these things does even better.
Sorry to breaktopic, but I'm looking into buying an 10gb IDE hard drive from IBM or maybe seagate. I'd split it between Linux and Windows. I remember hearing that Linux can't use more than a 4gb hard drive or something, Is this still true? I'm using RedHat 6.1 Does anyone have suggestions? Thanks!
Sorry to break topic, but.. I'm looking into getting a 10gb IDE Hd from IBM to split between Win9.x and Linux. I remember hearing soemthing that Linux can't use higher than 4.0 Gb, Is this still true? and is there anything else I should know? this is my first harddrive upgrade in 4 years. THanks, I'm using Redhat 6.1
If voting were effective, it would be illegal by now.
Wow. I really like these drives going faster and faster. I can't wait till they break the light barrier. Imagine a harddisk that spins faster than the speed of light. Due to time dilation, your requested data will arrive before you even requested it! That would be a solution to a lot of problems. Of course, there's the problem that it's impossible for matter to go faster than the speed of light, but hey, one has got to fantasize, right?
----------------------------------------------
the pun is mightier than the sword
Seagate should work on its harddrives quality before improving things. Their "buggy firmware" is still causing troubles, cf "{ DriveReady SeekComplete Error }".
>debug g=c800:5
:-(
:-)
if only that worked with my controller. It was one of the "newer" 16-bit ones, so that function wasn't supported. I had to search for MS-DOS software that still worked that would allow me to low level format the drive (no fun). These new-fangled graphical AMI BIOSes don't offer the services they used to...
>St252ax, so big you had to create 2 partitions, because of the 32MB partition limit imposed by, (you guessed it), Microsoft.
I was one of the lucky ones. Didn't stop using my C64 until DOS 4.01 was out.
If you could be told what you can see or read, then it follows that you could be told what to say or think - BoC
Would this be possible? With two independent sets of heads you could halve the latency, double the transfer rate, and reduce seek times. Simultaneous access to two different parts of the disk would speed up considerably.
Molly.
I have a 52X Kenwood, and lately it's been getting more picky about what CD's it wants to read. I'm also getting more and more buffer underruns while copying from that drive, so I have to copy through the HD. Never had that problem with my previous CD Reader. Seems like I've been reading reports on low reliability around the newsgroups also.
When it's working, it's a very fast drive! Too bad it doesn't rip audio as fast as my Sony CDRW though.
Now perhaps there's a firmware update, I just haven't gotten around to checking yet...
-CausticPuppy "Of all the people I know, you're certainly one of them." -Somebody I don't know
I work it out as 205 Hz, if 15,000 RPM is exact.
15,000 rpm. That's 500 rpm SLOWER than the redline of a Yamaha R6 motorbike.
HH
Yellow tigers crouched in jungles in her dark eyes.
She's just dressing, goodbye windows, tired starlings.
Once you have that set up, i'll come over and lipsink some Elvis for your entertainment
>HDD platters have abou the same stiffness as a pice of cardboard you
>lamebrained shithead.
Never cut your finger on a sheet of cardboard, I take it?
Well, if you were playing Madonna albums it was probably trying to commit suicide. :-)
-pf
Make affiliate bucks
Anyone know how much these go for on the street? Where to buy em?
It would be possiable to stick 4-8 of these in a RAID cage ah?
"`Ford, you're turning into a penguin. Stop it.'" -THHGTTG
First: 8008 + 5k Ram + front panel... who needs them keyboard and screen thingies.. :) Ah back when it took a real man to compute at home.
60 feet per second seems to be right. But if you convert that to 230 kph, then your feet is over 1 m long...
69 km/h is the correct peripheral speed. Still rather fast, but not as deadly as you said. Note also that if the disk breakes and runs through the case, then it must be a fairly large piece that breakes away. And if it's a large piece, then the speed will be lower, since some of it would come from closer to the center.
And most of the energy will be lost to get through the case (if it will at all). And after that (when the disk must be pretty deformed) it will have to get through a couple more layers of metal before it gets out of the case.
I'm not the least afraid.
You freakin' bone weasle. Get a life.
I think it's in the jargon file; I just remember it as a bit of folklore :)
... !berkley!prime3!hawk@olivetti.atc
:)--once he saw it, he could send a message right back up the same path. If he didn't, maybe you'd be lucky and someone else would see it; maybe not.
Usenet used to rely on the arpanet backbone where available, but most sites got their feed through modems. Sending email (off arpanet) required knowing not only the destination address, but the path of every machine that the message would hop along the way (but this was easy if responding to a post; just send it back from whence it came). To email me from back east you would have sent to
something like
!lilcompanyvax!decvax1!decvax5!
gad, it's been a while; maybe I have that in the wrong direction,
and I don't remember the names of the machine, but I think that
was my final address. i
Oh, and of the 30 or so newsgroups at the time, it seems to me that two were devoted to finding paths to people. Basically, a lot of posts like, "Does anyone have a path to George Jones at Olivetti in Cupertino?" If George knew your were looking for him, he would read those newsgroups until he found your message (or grep the newsspool
Anyway, I was saying that most sites got it through modem. Then there
were the sites that didn't, which got it by tape (Australia?), leading
to the observation,
"Never underestimate the bandwith of a [station wagon|747] full of
nine track tapes."
or something like that.
/end{reminisce}
I don't know why it only occurred to me in the last few days instead of years ago, but . . .
Density has gotten high, but if you want to hit two or three drives at once (swap, usr, tmp, home), you still need multiple drives. What about building a drive for which the groups of platters were separately accessible for different blocks of heads? You might even make it configurable--three groups of head-steppers, and a platter could attach it's heads to any of them?
This clearly wouldn't be a solution for servers, but it would seem to offer some huge benefits for workstations.
I've done some reading about these new drives and word is that the super high RPMs generate so much heat that Seagate has taken to routing coolant through the drive heads just like blood through your body. Of course the little coolant pipes are tiny tiny, but they're there, chillin out, to use the parlance of our times.
This is ridiculous. There are no cooling devices on HDDs. Drive HDAs are designed to run as hot as 65C indefinitely (and much hotter for short durations), which is easy to maintain with fans. Drive heads are solid-state devices which have very high heat tolerances.
Drive manufacturers have kept the power (and thus heat) dissipation down by reducing the diameter of the platters. The power consumed by the spindle motor is a fifth-order function of the diameter of the platters, so even a small reduction in size greatly reduces the spindle motor power requirements. Over the years, the electronics on the PCB have migrated to lower voltages, which also helps the power.
Why would the AC mains frequency have anything to do with the platter speed? Hard drives don't run directly off of the AC, they use the DC from the power supply.
I mean, what more can you say? you ROCK! This reminded me of abusing CD's in another way. When I went to school and worked at the campus newspaper, we got a LOT of music cd's for "review." It was cool when you got good ones but there was inevitably a lot of crap. I mean you put it on and listend to three songs, everyone either agreed it sucked or some poor soul took it home. Then we went to the parking lot after the press ran and had "the CD Discus Olympics (open division)". Cd's from the local pressing plant were extremely light and cheap and if you threw 'em high, they would shatter into a goodly number of pieces upon impact with the asphault..... :)
---
DO NOT DISTURB THE SE
We're still waiting...deeply buried in their Web site (or maybe not) is a quiet mention that the drives are expected out Q1 or Q2 of 2000.
So, it's nice to see the advance to 15KRPM, but this doesn't mean you'll be able to buy one anytime soon! Seagate seems to savor the big announcement about new tech waaaayyyy in advance of when you will actually be able to buy it.
Now, back on track, 6 ms isn't a bad seek time in that case, but I do beleive they quoted 3.9 ms in the article, or, 11.7 ms.
The average seek time is the sum of an average actuator movement, and an average wait for the disk to rotate to the start angle.
15000 rpm => 4 ms per revolution => average time to position the disk = 2 ms (when doing random reads).
They say that they have a random seek time of 3.9 ms => average actuator positioning time = 3.9 ms - 2 ms = 1.9 ms. The AC you commented aproximated this by 2 ms.
The average actuator movement is 1/3 of full stroke. (If you do some linearisations.) That makes the full stroke 3*1.9 ms = 5.7 ms. (Which was aproximated to 6 ms in last post.)
Clear enough?
That might be good (by my memory, it is), but that still isn't the sort of improvement I'd like to see. Get it down to 2 or 3 ms, then I'll be real happy. That would shorten the quake 3/unreal tournament loading times by 1/2, at least!
Well how would you know? You couldn't even do the simple calculations above, so why do you think you can predict what difference it would do to Q3/UT loading time?
I had a big long flame ready (your last comment pissed me off so much), but instead, I'll just make you look foolish:
My previous statement:
>That would shorten the quake 3/unreal tournament loading times by 1/2, at least! And fscking a hard drive wouldn't take such an unweidly amount of time!
Question:
>Well how would you know? You couldn't even do the simple calculations above, so why do you think you can predict what difference it would do to Q3/UT loading time?
Answer: Because, while my math abilities are not ubercool, I use common sense. While 1/2 isn't correct by a mile, it is used in this case as slang for "faster". Is english your mother tongue? If it isn't, I'll forgive you for that slip up. And I'll assume you agree fscking a hard drive is affected by seek time (so somehow I can be right on that but not the other? huh?)
Do you think that head seek time makes no difference to loading software? If you do, well, time to stop taking math, and get back to reality (try some technical courses).
If you could be told what you can see or read, then it follows that you could be told what to say or think - BoC
Slashdot a way of bringing out the worst in people:
I post something innocent like:
Fast spin time is nice... But how fast do the HEADS move? That's the other most important side of the equation...
And it becomes a pissing contest. See the replies:
>Try reading the article instead of trying to get a fast post (#6 in your case)
then I say (stupidly joining in the pissing contest):
>might I reccomend a speed reading course?\
and the contest continues:
>That's Alt-F. Oh, wait. You were using Windows, right?
Of course, I would be having a bad day, and drivers' reply to my remark about speed reading was reasonable in hindsight:
>Maybe speed reading is your problem. You don't go for comprehension.
But I disagree that average seek time is a realistic way to measure HD armature movement, and since that is all the article uses to measure the head speed, I don't find a lot in there useful.
Of course, I would have to shoot off my mouth and make my post nasty by quoting every fact in the article. Again, if I had realised just how friggin' vigilant slashdot users can be about what they think is right, I wouldn't damn well speak here at all. Usenet is less forboding, for chrissakes.
Yeah, and I ignored the nice reply (#177) to my comprehension flame, which was sticking to the topic. Although they still use seek time for a comparison, which as is admitted by all here, a combination of both rotation speed and seeking speed (which, as I said, I don't think mix for a good stat).
Yeah, and I did say vaguely defined things like seek times. So I got a nice post explaining it to me. No problem, and I figured that their post seems like a reasonable explanation. As I figured, the rotational speed is part of the equation, messing up true head seek time.
A reply to this also explains how new smaller diameter platters are increasing seek times.
But then someone with a bad attitude has to get the math absolutely perfect, never mind the fact that we are all settled on the fact that seek time != head speed. That is like spell checking someone's post and pointing out every error. That's being a jerk. Then they turn into an asshole by saying:
>Clear enough?
What, do you think I'm an idiot? If someone said that to me, and I was their boss, I'd fire their ass.
>Well how would you know? You couldn't even do the simple calculations above, so why do you think you can predict what difference it would do to Q3/UT loading time?
I'd offer to give them their job back if they could prove that a 10 ms drive is going to load Q3/UT faster/same speed as a 4 ms drive. Needless to say, that comment wasn't at all thought through.
By this time, I am pissed off, so I make that asshole eat his own words.
Then you, dickwad, say:
>You are being an idiot.
>You may have "PROVED" you read the article, but you certainly didn't *understand* it
>Despite your whining
>The very quote you mocked with your ignorant "sarcastic" comment
>A final clue...
>they don't design these high end drives so you can load a fucking UT level fast.
>I think they just might have a better idea of what to do to fulfill that need than you do
If you think any of those comments were necessary, say so now. Otherwise, realise that saying things like that means war.
If you were to make sure that when someone makes a mistake, you were to always one up them in life by being sure to point out at every chance you get that they are an asshole, you'd eventually find someone that would put their fist in your face.
Well, here it is, virtually. I've picked apart all the replies to my post, and commented on my mistakes, and looked at all the nasty replies, and boy oh boy, you certainly have won. Yours is the nastiest yet.
Do you think you can do even better next time?
If you could be told what you can see or read, then it follows that you could be told what to say or think - BoC