Slashdot Mirror


Cray's New Solid State Storage

Sivar writes: "Cray, a well known vendor of extremely fast supercomputing hardware, has introduced a storage system with a 224 GB capacity. The large size seems impressive, but the device can also transfer an unprecedented 80GB(!!) every second. That's more bandwidth than the main memory of most servers, and it's just for storage. For comparison's sake, a typical dual channel DDR motherboard has a bandwidth capacity of barely 4.2GB/sec." Yow.

355 comments

  1. Wow, that is utterly amazing. by TheDick · · Score: 1

    I love logging into /. and seeing a headline that makes my draw jop. Its really cool.

    I want one.

    --

    1. Re:Wow, that is utterly amazing. by NanoGator · · Score: 1

      "But hell, if I want to see some titties, I'll go to the fucking nudie bar and throw some money around. Those sluts will do anything for two hundred bucks."

      I wondered if Bud Bundy posted here. Heh.

      --
      "Derp de derp."
    2. Re:Wow, that is utterly amazing. by Anonymous Coward · · Score: 0

      Want your jaw to drop further? Look here

    3. Re:Wow, that is utterly amazing. by Anonymous Coward · · Score: 0

      I thought it said "Crazy new storage"!

    4. Re:Wow, that is utterly amazing. by Anonymous Coward · · Score: 0

      These are some great troll.

      Keep it up, guys!!

      I think trolling is funny, myself

    5. Re:Wow, that is utterly amazing. by Anonymous Coward · · Score: 0

      you just want your dick sucked ... yahhooray for faggots like you..

      -

  2. Hole E Shyte... by CrazyDuke · · Score: 0, Troll

    Need I say more? I want one!!!!

    --
    Any sufficiently advanced influence is indistinguishable from control.
  3. obligatory by Anonymous Coward · · Score: 0

    imagine these in RAID 0!

    1. Re:obligatory by (outer-limits) · · Score: 1

      I am going to buy 10 for my boewolf cluster.

      --

      Microsoft - Where would you like to go today, Maybe Jail?

  4. Super storage, super price. by Keighvin · · Score: 2, Insightful

    This is more of a conglomeration of current technology into a pricy solution, not so much a stellar advance.

    --
    Any spoon would be too big.
    1. Re:Super storage, super price. by doooras · · Score: 5, Funny

      yeah, and computers are just arrangements of existing silicon, copper, aluminum....

    2. Re:Super storage, super price. by zmooc · · Score: 1

      True. But which technology?

      --
      0x or or snor perron?!
    3. Re:Super storage, super price. by ClickNMix · · Score: 1
      This is more of a conglomeration of current technology into a pricy solution, not so much a stellar advance.

      Perhaps true, however, it does mean if you have that money, and need that power, its now yours for to buy. It might not be anything new in principle at its barest level, but then, what is.

      --
      I saw the light at the end of the tunnel... But it was just someone with a flashlight bringing more work.
  5. PThe press Release.... by Phosphor3k · · Score: 1

    Does not say how fast it can write to it's storage medium. I'd be very interested to know.

    1. Re:PThe press Release.... by Anonymous Coward · · Score: 0

      More or less the same as the read bandwidth, 80 GB/sec (according to the press release). The limit is the channel, not the device itself.

    2. Re:PThe press Release.... by billstr78 · · Score: 4, Informative

      It might use a technology known as MEMS which is probe based storage. Probe-based storage system supports probe-based reading and writing of bits, is based on non-rotating media and initially
      expected to support storage densities on the order of 100 to 300 Gbit/inch2. The storage
      devices are envisioned as two rectangular sleds, one with storage media and the other
      with a sparse array of very small read-write heads, in the range of thousands to millions.
      Seeks will require x and y motion of one of the sleds relative to the other. These devices
      are intrinsically highly parallel because some or all of the heads will be able to operate
      simultaneously.
      [MEMS Modeling]

    3. Re:PThe press Release.... by Anonymous Coward · · Score: 0

      Mirror it, or RAID it, fail it over, just make it redundant and persistent, and this IS the storage medium. Fastest SCSI on the market is 320 Mbytes/ sec.

    4. Re:PThe press Release.... by Anonymous Coward · · Score: 0

      >It might use a technology known as MEMS which is probe based storage

      Nope, it's plain old SDRAM. Just lots of it, on multiple controllers.

    5. Re:PThe press Release.... by Anonymous Coward · · Score: 0

      Does not say how fast it can write to it's storage medium. I'd be very interested to know.

      Writes will max out at 110bps. This is because it requires several trained technitions to actually write out the data will very small pens.

    6. Re:PThe press Release.... by Anonymous Coward · · Score: 0

      It's unlikely that a MEMS-type technology is used, as i think some mechanical laws would limit the rate of data transfer. Although MEMS devices are potentially faster than conventional devices like hard drives, they still have moving parts, however small, which would limit their transfer rate to below that level. It seems more likely that the system is based on some sort of extremely high-response SRAM; an expensive solution, but probably the best conventional technology has to offer.

  6. Hardware review by mESSDan · · Score: 5, Funny
    C'mon, where are the obligatory Quake3/UT FPS statistics? I want to know if I'm going to get 1,000,045 FPS or 1,000,0053 FPS. Don't they read Tom's Hardware?

    Heh

    --

    -- Dan
    1. Re:Hardware review by Sivar · · Score: 2

      Games usually run from main memory, so the storage system would only effect load times. Of course, your Q3 frame rates would likely shoot up if you ran the game from your swapfile...

      --
      Computer Science is no more about computers than astronomy is about telescopes. --E. W. Dijkstra
    2. Re:Hardware review by JS_RIDDLER · · Score: 1

      I'm Sold... Where can I buy or build one, and how much!?

      --
      _JS
    3. Re:Hardware review by psamuels · · Score: 1
      C'mon, where are the obligatory Quake3/UT FPS statistics? I want to know if I'm going to get 1,000,045 FPS or 1,000,0053 FPS. Don't they read Tom's Hardware?

      I'd like to see it compile a Linux kernel in, say, 0.04 seconds. Too bad neither Quake3 nor gcc are particularly well-suited to vector processing (aka SIMD, if I'm not mistaken on my terminology).

      --
      "How can you claim that you are anti-crack, while still writing a window manager?" — Metacity README
    4. Re:Hardware review by RevDobbs · · Score: 1

      Just send your first born, thank you. And to ensure prompt payment, why don't you leave your "joy stick" hand as a deposit...

    5. Re:Hardware review by realdpk · · Score: 4, Funny

      Yeah, and if it was on Tom's Hardware, they'd include some practically useless commentary on how it is 5% faster if you overclock it. Woo hoo!

      :)

      (on that note, is there a hardware site out there that does not have this bizarre overclocking "bent"?)

    6. Re:Hardware review by keiferb · · Score: 2, Funny

      True, but that 5% would be equivalent to a couple of pretty beefy desktop machines in this case. =)

    7. Re:Hardware review by neoform · · Score: 2, Funny

      imagine you had a monitor capable of displaying those frame rates... i'd imagine you would be bald in 3 minutes..

      --
      MABASPLOOM!
    8. Re:Hardware review by Anonymous Coward · · Score: 0
    9. Re:Hardware review by (outer-limits) · · Score: 1

      Just don't let Tom turn off the cooling fans. I can see the smoke pouring out now.

      --

      Microsoft - Where would you like to go today, Maybe Jail?

    10. Re:Hardware review by pmsr · · Score: 1

      5% faster and it that it just crashed twice a day. Very stable they will say.

      /Pedro

    11. Re:Hardware review by Anonymous Coward · · Score: 0

      quake3 is not designed for a cray, it would suck ass.

    12. Re:Hardware review by Anonymous Coward · · Score: 0

      (on that note, is there a hardware site out there that does not have this bizarre overclocking "bent"?)

      ars

    13. Re:Hardware review by Ctrl-Z · · Score: 1


      Hmmm... I find your ideas intriguing and I would like to subscribe to your newsletter.

      --
      www.timcoleman.com is a total waste of your time. Never go there.
  7. Wow by dspeyer · · Score: 2, Interesting

    I'm impressed.

    Of course, it probably won't work on ordinary computers (after all, sticking that onto a SCSI bus would be sort of a waste), but eventually we'll get our hands on this stuff.

    Anybody dare to ask how much it costs?

  8. Money by Warped-Reality · · Score: 1

    I'm sure it will be out of reach of the average /. geek for a while... but it looks really cool

    --
    This is not the greatest sig in the world, no. This is just a tribute.
  9. slashdot effect by pangu · · Score: 1

    now we just need to hope that their webpage is served up by one of these systems...

    1. Re:slashdot effect by myz24 · · Score: 4, Funny

      I can hear it now...

      /.: "What kind of server do you run?"

      C: The new super duper Cray with the new 224 gig storage system moving data around at 80GB!!

      /.: "Whats your connection to the net like?"
      C: 256kb DSL line, why?

      /.: "...."

    2. Re:slashdot effect by Anonymous Coward · · Score: 0

      Let's see:
      Natalie Portman - 20, cute
      Madhuri Dixit - 30, stinky ass paki.

      Hmm...

  10. yey tivo supreme by rtstyk · · Score: 1

    Why would I possibly want one? This is one of those too cool to be useful for mortals kinda things...

    But wait! What if someone used it to write real time HDTV format to the disk on this? Maybe they should come up with super duper Tivo box? That would be cool...

    d.

    --
    I hate the fact that you people don't salute me
    1. Re:yey tivo supreme by psamuels · · Score: 1
      But wait! What if someone used it to write real time HDTV format to the disk on this? Maybe they should come up with super duper Tivo box? That would be cool...

      Yeah, a Tivo that costs somewhere in the 5-figure range, uses several kilowatts of 3-phase power, is water-cooled, takes up half your living room, hums with satisfaction, and isn't anywhere near the shielding requirements of a class B computing device. But the Blinkenlights are groovy, no doubt. And oh yeah, it can record thousands of HDTV channels simultaneously.

      --
      "How can you claim that you are anti-crack, while still writing a window manager?" — Metacity README
    2. Re:yey tivo supreme by s390 · · Score: 2

      Maybe they should come up with super duper Tivo box? That would be cool...

      Think media streaming server. Compaq and Sun are already building these, but it sounds like Cray might beat them easily, if they bother to build one.

    3. Re:yey tivo supreme by -Surak- · · Score: 1

      5 figures? I'd be really surprised if it's that cheap. I'd tend to think that something like would be WELL into the multi-hundred-kbux - and that's just for the storage we're talking about in this article. Hell, I'd be surprised if the power supply alone is less than 5 figures.
      Not that I'm criticising Cray... outfits that need this can afford it by definition - the technology will trickle down to the rest of us eventually.

  11. This Has to be Said... by Anonymous Coward · · Score: 0

    Imagine a beowulf cluster of these! :-)

    1. Re:This Has to be Said... by Anonymous Coward · · Score: 0

      A Beowulf cluster of Harddisks?

      Wow the uses?

      A Beowulf cluster of Crays? Strange little spammertroll.

  12. Obligatory... by guiding_knight · · Score: 1

    I'd say imagine a beowulf cluster of those, but...

    a) I dont want to get smacked by someone

    b) That thing puts the capacity of a beowulf cluster to shame...

    Anyway, awesome piece of technology. Any drawbacks, anyone?

    --
    LOTR: Elijah Wood is a munchkin asshat. Yes, asshat. LOL.
    1. Re:Obligatory... by Anonymous Coward · · Score: 0
      Any drawbacks, anyone?

      Umm...

      • Price of the unit
      • Price of the computer to connect to
      • Price of the electricity
      • Price of the chilled air
      • Price of the admin
      Drwabacks ? .. no, not really.
    2. Re:Obligatory... by Anonymous Coward · · Score: 0

      Any drawbacks, anyone?

      mozilla still takes too long to compile.

    3. Re:Obligatory... by Sivar · · Score: 2

      You wouldn't put them into a beowulf cluster anymore than you would put hard drives into a beowulf cluster. RAID is what you want, but really, a RAID array of these things would be pretty rediculous. :-)

      --
      Computer Science is no more about computers than astronomy is about telescopes. --E. W. Dijkstra
    4. Re:Obligatory... by pixel.jonah · · Score: 1

      if you were looking for shared storage for your 1000 node cluster - yeah... You'd need a separate gigabit network just for your storage...

    5. Re:Obligatory... by Anonymous Coward · · Score: 0

      > a RAID array of these things would be pretty
      > rediculous. :-)

      I couldn't disagree more!
      However, a RAID array of these things would be pretty ridiculous. :-)

    6. Re:Obligatory... by Anonymous Coward · · Score: 0

      well, beowulfs are for more than just shared data storage. The problem is, networking these buggers effectively would take a LOT of money.

  13. Hmmm... by Stinky+Boy · · Score: 3, Funny

    None on EBay yet...

    Well, looks like I'll have to wait a few weeks.

    --
    Ceci n'est pas une sig.
  14. somebody do the math for me cause im lazy by KaizerWill · · Score: 1

    how many days worth of mp3s is that?

    1. Re:somebody do the math for me cause im lazy by Stinky+Boy · · Score: 1

      Assuming 128Kbps...
      And assuming there's two bits of overhead per byte...
      And assuming then that the average song is 3 minutes long (that's average airplay songs for you - if you like Iron Butterfly, go with 17 minutes in your calculation)...

      That's (3*60)*(128/10) = 2,304 * 1,024 = 2,359,296 bytes per song.

      Then, 226,000,000,000 bytes (assuming 1,000,000,000 bytes per GB, since we're talking a hardware vendor here...) divided by (2,359,296) =

      95,791 whole songs available instantly.

      --
      Ceci n'est pas une sig.
    2. Re:somebody do the math for me cause im lazy by nytes · · Score: 1

      And at 80 GB/sec., you'll be able to listen to all 95,791 of them in 3 seconds.

      --
      -- I have monkeys in my pants.
    3. Re:somebody do the math for me cause im lazy by swg101 · · Score: 1

      finishing the math:

      at 3min. per song, + 3second leader in between...

      roughly 202 days of music straight through!

      that's not too shabby, I just want to know how you will download that much music.

      -keep the faith, Mozilla 1.0 is almost here!

      --
      Like pi? Try 10,000 digits.
    4. Re:somebody do the math for me cause im lazy by GigsVT · · Score: 1

      Not as many days of MP3s as this.

      Since I built it, I did the math on the MP3s first thing. What else does one do when presented with an incomprehensible amount of space.

      Lets assume 2MB per minute, a little high for 128kbps, but lets assume you have 160s in there too.

      Server 1: 1.6TiB
      Server 2: 1.8TiB (I think the 1.9 on the page is a typo)

      That's 3.4 TiB, 3481GB, 3.5Mln MB rounded.

      divide by 2, 2mb per minute.
      divide by 60 minutes in an hour
      12 hours in a day
      365 days in a year.

      6.7 Years of Mp3s without a single repeat. For less than $17,000.

      If you assume 1MB/minute of MP3, which is closer to reality, it's over 13 years of MP3s.

      Read speed is 128MBytes/sec, write speed is 42MBytes/sec.

      You don't need a cray, you can build this at home, scaled down.

      --
      I've had enough abrasive sigs. Kittens are cute and fuzzy.
    5. Re:somebody do the math for me cause im lazy by krugdm · · Score: 2

      95,791 whole songs available instantly.

      Ad slogan: "The new iCray. 95,791 songs. In half your basement."

    6. Re:somebody do the math for me cause im lazy by Anonymous Coward · · Score: 0

      Mozilla 1.0 is ALREADY HERE! Great, innit! I'm amazed how much better it's got. Wow - imagine what Mozilla 2.0 would be like...

    7. Re:somebody do the math for me cause im lazy by QuaZar666 · · Score: 1

      And as we all know. IDE drives work great in high bandwidth vector based computers with over 1000 nodes needing a storage solution that is able to deliver a massive throughput.

      There is a reason why you might need 80 GB/sec and its not for mp3's.

      Qua

    8. Re:somebody do the math for me cause im lazy by GigsVT · · Score: 1

      I know, the guy asked about MP3s through. It was kind of a stupid question since these are obviously optimised for bandwidth and not space. That is why I showed him the IDE RAID.

      --
      I've had enough abrasive sigs. Kittens are cute and fuzzy.
  15. Uh Oh by Anonymous Coward · · Score: 0

    Here come all those lame "this'll barely hold my mp3 collection" jokes.
    Damnit.

    1. Re:Uh Oh by Anonymous Coward · · Score: 0

      This'll barely hold my mp3 collection...

      To get to the other side!

    2. Re:Uh Oh by Anonymous Coward · · Score: 0

      They're not jokes...

  16. Slashdotted???? by TheBracket · · Score: 2, Funny

    Please tell me that cray.com isn't slashdotted. Oh, how the mighty have fallen!

    --
    Lead developer, http://wisptools.net
    1. Re:Slashdotted???? by billstr78 · · Score: 1

      Maybe they are using cheap Wintel hardware to host the site, like M$ used freeBSD to host thier "We have the way out" site.

    2. Re:Slashdotted???? by Boone^ · · Score: 2

      Last I checked it was a Solaris box, before that an SGI box.

    3. Re:Slashdotted???? by Anonymous Coward · · Score: 0

      As of 7:25PM there not slashdotted...

    4. Re:Slashdotted???? by TheTomcat · · Score: 3, Interesting

      netcraft says it's solaris 8 and apache.

      Of COURSE they're not hosting the machine on a cray. That's complete overkill, even for Cray themselves. The electrical costs alone would be on par with a top-of-the-line hosting package (I imagine).

      S

    5. Re:Slashdotted???? by jacobcaz · · Score: 1
      OF COURSE they're not running the web site on a Cray, because a Cray isn't designed to run a web server (not effectivly).

      Unless someone knows how to write a web server using nothing but vector math! :-)

    6. Re:Slashdotted???? by Boone^ · · Score: 2

      We've gotten some open source apps compiled for the newer Cray boxes, but I'm not sure if Apache was one of them. Can anyone from Mendota comment?

      But yeah, the Cray boxes we use are used for simulations, not webhosting. :P

    7. Re:Slashdotted???? by zer0*ryok0 · · Score: 1

      maybe for just their own website

      but we could use this new storage device, and host the WHOLE INTERNET on one cray.

      --
      the only fact is that everything is an opinion
    8. Re:Slashdotted???? by TheTomcat · · Score: 1



      Yes, you're right.
      The WHOLE INTERNET could be contained in 227 gigs.

      </sarcasm>

      S

  17. Sweet... by llamalicious · · Score: 1

    continued refinement and technical advances in the computing field. That's the real meat I like to see on /.
    I can't wait for 2030 to see where technology ends up.

    I really do want to see some real information about the underlying design of the stuff.

  18. /.'ed by BlackSol · · Score: 4, Funny

    anyone else find it hilarious that the site is slashdotted?

    I know its probably hosted by someone else but come on just the idea that we slashdotted a cray is awesome :)

    --
    $sig=$1 if($brain =~ /idea\s+(.*)/i);
    1. Re:/.'ed by Usquebaugh · · Score: 1

      I was so bummed. It's true that Cray is a shadow of it's former self :-( Like SGI et al Are there no engineering companies left?

    2. Re:/.'ed by Jonny+Ringo · · Score: 1

      I know they should be able to handle all 80 gigs of us :-)

    3. Re:/.'ed by wedg · · Score: 2

      I know its probably hosted by someone else but come on just the idea that we slashdotted a cray is awesome :)

      Crays do vector computing. Why the hell would they have a lot of bandwidth? Though, the day we slashdot UUnet or AT&T I'll laugh until I pee myself.

      --
      Jake
      Dating: while( 1 ){ call_girl(); get_rejected(); drink_40(); } return 0;
    4. Re:/.'ed by jsse · · Score: 1

      /. a cray? I woudln't dare piss it up - I'm not sure I want to take 80G bits of reverse ping from it. :/

      J/K

  19. An interesting side effect... by Sivar · · Score: 4, Insightful

    Get one of these, downgrade your system to 8MB RAM, and run everything from swap...

    Watch your system's responsiveness double.

    --
    Computer Science is no more about computers than astronomy is about telescopes. --E. W. Dijkstra
    1. Re:An interesting side effect... by geekoid · · Score: 3, Insightful

      why use any RAM at all? That would be an interesting turn of events, wouldn't it? As a matter of fact, that might be an actual Paradigm Shift in computer architecture.

      --
      The Kruger Dunning explains most post on /. http://en.wikipedia.org/wiki/Dunning%E2%80%93Kruger_effect
    2. Re:An interesting side effect... by jbridge21 · · Score: 3, Insightful

      Too many context switches makes your system go dog slow. While the transfer rate is extremely high, the latency of talking to a piece of storage that is perhaps several feet away, at the speed of light, is too high.

    3. Re:An interesting side effect... by Anonymous Coward · · Score: 0
      Get one of these, downgrade your system to 8MB RAM, and run everything from swap...
      I just want to know what the speed of the memory on the computer connecting to this disk storage would be! I am betting that it ain't 4.2GB/s. The system probably has main memory running at 100's of GB per second in terms of bandwidth. It probably has a 2048-bit or more bus as well, but hey, that is details. Fact is, it should be faster than the disk storage... Graham
    4. Re:An interesting side effect... by Elwood+P+Dowd · · Score: 2

      Um, not with any current OS or standard PC hardware. If it treated the SSD as a hard drive, it would insist on transfering everything into those 8MB before operating on it. So you'd beat the hell out of your 8MB of RAM, and you'd be doing it at RAM speeds, not 80GB/sec. In this completely hypothetical setup, it would definitely be faster to have more RAM. No PC could probably connect to this device, let alone exercise it.

      --

      There are no trails. There are no trees out here.
    5. Re:An interesting side effect... by John_Booty · · Score: 1

      Dude... I have mod points. I almost modded you down just for using the phrase "paradigm shift". :)

      I agree with you though, btw.

      --

      OtakuBooty.com: Smart, funny, sexy nerds.
    6. Re:An interesting side effect... by teslatug · · Score: 1

      Electrons are not photons...they have mass and thus cannot travel at the speed of light even in vacuum...in a wire there is not enough room for them to travel at a high speed for long as they will collide...thus their mean speed is significantly reduced (wires have resistance).

    7. Re:An interesting side effect... by (outer-limits) · · Score: 1

      I have been waiting for the end of this RAM/DISK pain in the a** for years. IBM AS/400s have had an io less architecture for years. Everything is paged. I look for the day when I don't have to 'save something to disk'. All objects will have a class of persistent, and then you are done. And there is no penalty in terms of speed for doing so. In the background, the garbage collector would be getting rid of everything that hasn't been referenced in the last ten years.

      --

      Microsoft - Where would you like to go today, Maybe Jail?

    8. Re:An interesting side effect... by inburito · · Score: 2

      It is actually a lot more complex than that. You're thinking that a single electron would convey the message and have to move the distance between transistors in order to do this. This is however not the case.

      Knowing the basic principle I read the following analogy somewhere:

      Consider a fairly long empty garden hose. Once water is first turned on it takes a while for any water to come out of the other end of the hose. But, once the hose is filled with water any subsequent attempts to change the pressure, or turn water on and off, result in a rather rapid response over that same length.

      So, it is pulses that convey this "information" in the water hose and similiarly in electronic chips electronic pulses of this very same kind are used to transmit information across the chip. These signal pulses move substantially faster than the individual electrons meaning that the individual speeds of electrons are in the end rather insignificant but still for the actual information transfer the limiting factor is that of speed of light..

    9. Re:An interesting side effect... by teslatug · · Score: 1

      I believe that the limiting factor is actually thermal velocity. Electrons move much more quickly on their own, if you look at the aggregate movement it is actually much slower.

    10. Re:An interesting side effect... by O_Sleep · · Score: 1

      Is this system copper based? Why wouldn't it be fiber based?

    11. Re:An interesting side effect... by Merk · · Score: 3, Interesting

      That's the way PalmOS computers work today. There is no difference between your short-term storage and your long-term storage. That's why, with a Palm application, when you enter an event or click a checkbox, you don't have to "save" the results, because it's all in the same memory area.

    12. Re:An interesting side effect... by jbridge21 · · Score: 1

      Your corrections only make my argument stronger. I was going for the theoretical best case, and saying that it wouldn't work even then.

  20. Transfers more than it can store... by IsleOfView · · Score: 2

    Anyone find this comment from the release kind of ironic?

    The field-upgradeable SSD system can hold 27 copies of the Human Genome and transfer data at a rate equivalent to 100 Human Genomes per second.

    yum!

    1. Re:Transfers more than it can store... by taniwha · · Score: 4, Funny

      and transfer data at a rate equivalent to 100 Human Genomes per second.

      Yeah but then so can a fully laden school bus ....

    2. Re:Transfers more than it can store... by Anonymous Coward · · Score: 1, Funny

      I could do better than that with an ordinary catapult and a supply of testes.

      SPLAT!

    3. Re:Transfers more than it can store... by MarcoAtWork · · Score: 2

      I find it wrong, the numbers don't add up...

      If the unit can store 224 gigs and transfer 80gigs/sec, it means that it theoretically takes nearly 4 seconds to transfer all of its data.

      If it contains 27 human genomes, I find it hard to believe that it can transfer 100 human genomes/sec (4x) it should be able to transfer 6.5 human genomes/sec (1/4)

      Methinks whomever wrote this press release instead of pressing the '/' key on their calculators pressed '*'

      --
      -- the cake is a lie
    4. Re:Transfers more than it can store... by d0s · · Score: 1

      not if it's the short bus...

      /me runs

    5. Re:Transfers more than it can store... by edrugtrader · · Score: 2

      your reference to 'gigs' might be Gb in one case and GB in another.

      --
      MARIJUANA, SHROOMS, X: ONLINE?! - E
    6. Re:Transfers more than it can store... by CityZen · · Score: 1

      It looks like they were off by a decimal place for the "100" figure. It should be 10 human genomes/sec for the transfer rate. This works out if you do the proportion for 224:80 to 27:x.

  21. If only ... by Anonymous Coward · · Score: 1, Informative

    ... they were running off of one.

    SEATTLE (April 8, 2002) - Cray Inc. today announced the availability of the world's highest-capability expansion memory and data transfer system. The new system, compatible with Cray SV1e(TM) and Cray SV1ex(TM) supercomputers, includes a 224-gigabyte Solid State Disk (SSD) with a data transfer rate of 80 gigabytes per second-800 or more times faster than the 10- to 100-megabytes/second speeds typical with today's disk servers.

    The field-upgradeable SSD system can hold 27 copies of the Human Genome and transfer data at a rate equivalent to 100 Human Genomes per second. With their 32-gigabyte central memories and the new SSD system, Cray SV1(TM) series supercomputers now provide up to a quarter terabyte of ultrafast memory. They can also be linked to a virtually unlimited number of standard disk servers for additional capacity, and to other computer systems via high-speed networking.

    "With the new SSD system, Cray SV1 series supercomputers can handle extremely large, data-intensive problems with unprecedented speed, convenience and cost-effectiveness," said Jerry Loe, Cray vice president of worldwide sales and service. "This will be particularly useful in bioinformatics, and for complex automotive and aerospace applications." The Cray SV1 series, named "Best Supercomputer" in 2001 by the readers of Scientific Computing & Instrumentation magazine, includes special hardware features for bioinformatics.

    "With the new SSD, bioinformaticists will be able to work with several copies of the Human Genome at a time, or perform whole genome comparisons, or pursue drug design and discovery, without wasting valuable compute time waiting for standard disk data transfers," said Jef Dawson, Cray's manager of bioinformatics development and marketing. "The SSD can keep up with the Cray SV1 parallel supercomputers' processors, which perform up to 12 operations per clock cycle."

    Dawson said the SSD will benefit virtually any application requiring large data sets. "The popular automotive application MSC/Nastran ran 2.5 times faster using the new SSD capability. Applications that run 'out of core,' including the popular GAUSSIAN chemistry codes, are also well suited to the SSD. You can think of the SSD as the world's biggest cache memory, or the world's biggest I/O buffer. Either way, it offers the world a new capability."

  22. You could store alot of pr0n on this. by Shut+the+fuck+up! · · Score: 1

    There, it's been said.

  23. Extra storage? by Indomitus · · Score: 1

    They can also be linked to a virtually unlimited number of standard disk servers for additional capacity,

    Yes, just in case you need extra capacity for your 224 gigabyte solid state harddrive. Sheesh.

    I also like how they're using 'copies of the Human Genome' for their storage metaphor instead of 'copies of the Library of Congress'.

    1. Re:Extra storage? by glwtta · · Score: 5, Funny

      hell, I can fill up 224 gigs - people actually doing useful things with it certainly will be able to.

      --
      sic transit gloria mundi
    2. Re:Extra storage? by Anonymous Coward · · Score: 0

      Yes, just in case you need extra capacity for your 224 gigabyte solid state harddrive. Sheesh.

      uhh... You're kidding, right? To people who own crays, 224 gigs is *NOTHING*. They talk in terabytes.

      Hell, even I have 170 gigs on my box.

      Though 224 or 550 gigs could be a good cached section of a database.

  24. Can you imagine... by Akardam · · Score: 1, Redundant

    ... a Beowulf cluster of these?

    *ducks rotten tomatoes thrown by rabid /.ers*

    I'm sorry, but someone had to say it!

  25. How the hell do you slashdot a supercomputer? by shyster · · Score: 1, Redundant
    Wow. The /. effect is at it again. 10 comments, and Cray's (maker of "the world's most powerful supercomputers") website is barely whimpering along. After 5 server timeouts, I finally got it to load. Maybe Cray should outfit their webservers with the new storage! Here's the text for those that come after me:

    CRAY INC. OFFERS WORLD'S HIGHEST-CAPABILITY EXPANSION MEMORY/DATA TRANSFER SYSTEM

    System Can Transfer 100 Copies of Human Genome Per Second and Quickly Handle Other Very Large, Data-Intensive Problems

    SEATTLE (April 8, 2002) - Cray Inc. (Nasdaq NM: CRAY) today announced the availability of the world's highest-capability expansion memory and data transfer system. The new system, compatible with Cray SV1e(TM) and Cray SV1ex(TM) supercomputers, includes a 224-gigabyte Solid State Disk (SSD) with a data transfer rate of 80 gigabytes per second-800 or more times faster than the 10- to 100-megabytes/second speeds typical with today's disk servers.

    The field-upgradeable SSD system can hold 27 copies of the Human Genome and transfer data at a rate equivalent to 100 Human Genomes per second. With their 32-gigabyte central memories and the new SSD system, Cray SV1(TM) series supercomputers now provide up to a quarter terabyte of ultrafast memory. They can also be linked to a virtually unlimited number of standard disk servers for additional capacity, and to other computer systems via high-speed networking.

    "With the new SSD system, Cray SV1 series supercomputers can handle extremely large, data-intensive problems with unprecedented speed, convenience and cost-effectiveness," said Jerry Loe, Cray vice president of worldwide sales and service. "This will be particularly useful in bioinformatics, and for complex automotive and aerospace applications." The Cray SV1 series, named "Best Supercomputer" in 2001 by the readers of Scientific Computing & Instrumentation magazine, includes special hardware features for bioinformatics.

    "With the new SSD, bioinformaticists will be able to work with several copies of the Human Genome at a time, or perform whole genome comparisons, or pursue drug design and discovery, without wasting valuable compute time waiting for standard disk data transfers," said Jef Dawson, Cray's manager of bioinformatics development and marketing. "The SSD can keep up with the Cray SV1 parallel supercomputers' processors, which perform up to 12 operations per clock cycle."

    Dawson said the SSD will benefit virtually any application requiring large data sets. "The popular automotive application MSC/Nastran ran 2.5 times faster using the new SSD capability. Applications that run 'out of core,' including the popular GAUSSIAN chemistry codes, are also well suited to the SSD. You can think of the SSD as the world's biggest cache memory, or the world's biggest I/O buffer. Either way, it offers the world a new capability."

    For more information on the Cray SV1ex series, visit www.cray.com or contact your local Cray sales representative.

    Don't forget the legalese on forward looking statements, and registered trademarks as well.

  26. I wonder... by PrimeWaveZ · · Score: 1

    I wonder if this would work with my TRS-80...

  27. New Measurement System? by MikeyLikesIt! · · Score: 4, Insightful
    The field-upgradeable SSD system can hold 27 copies of the Human Genome and transfer data at a rate equivalent to 100 Human Genomes per second.

    I guess that using standard measurements (GB and GB/sec) just isn't intuitive enough! But why use the humane genome as a reference? Is that REALLY more intuitive to most people? Does anyone (besides geneticists) really understand how much information is in the human genome?

    --

    I dunno... What do you wanna do?

    1. Re:New Measurement System? by glwtta · · Score: 3, Funny

      that is the absolute worst reference "unit" I've ever seen! Depending on how you define "human genome" it's anywhere between several hundred MB to several hundred GB. Can we just go back to the more reliable and accepted units, like the LOC (Libraries of Congress)?

      --
      sic transit gloria mundi
    2. Re:New Measurement System? by schmink182 · · Score: 1

      Just out of curiosity, how many copies of the human genome would anyone currently be trying to store on one computer at this time? I imagine there aren't very many decoded genomes out there, so would people want to copy it 10 or 20 times in case they need a couple of backups? And if they only have 10 or 20 copies, why (especially since they only have 10 or 20 copies) would they want to send 100(!) of these genomes to someone else?

    3. Re:New Measurement System? by curunir · · Score: 2

      Is that REALLY more intuitive to most people?

      If this were a consumer device, I'd probably agree with you. But who, besides geneticists, is gonna see one of these anytime soon?

      --
      "Don't blame me, I voted for Kodos!"
    4. Re:New Measurement System? by teg · · Score: 2

      They couldn't care less what "most people" think - they're selling to a small, sophisticated set of users which pay to have the performance from the future today. I'm a little surprised they mentioned any scale whatsoever...

    5. Re:New Measurement System? by pboulang · · Score: 2, Funny

      More reliable and acceptable units would have to be in football fields. Of course, I measure my velocity in nano-parsec per micro-fortnights, but that's not a bad unit. I suppose for measuring the number of bits, one would need a good conversion for surface area as in GB per sq in... thus 240 GB could be construed as something like 2 micro-football fields.

      --

      This comment is guaranteed*

      *not guaranteed

    6. Re:New Measurement System? by MrDelSarto · · Score: 1

      I prefer the unit of the "Moby Dick" (MD) -- as in "this new optic fibre link could transfer 5000 copies of Moby Dick every second".

    7. Re:New Measurement System? by rgmoore · · Score: 2
      But why use the humane genome as a reference? Is that REALLY more intuitive to most people? Does anyone (besides geneticists) really understand how much information is in the human genome?

      Of course most non-biologists don't really understand just how big the human genome is. That's why they're using it as a reference. The genome is actually smaller than many people think (about 3 GB at one base per byte, but trivially compressible to 1/4 that), which means that expressing things in terms of the genome is a good way making your equipment sound more impressive than it actually is.

      --

      There's no point in questioning authority if you aren't going to listen to the answers.

    8. Re:New Measurement System? by rgmoore · · Score: 4, Informative
      Just out of curiosity, how many copies of the human genome would anyone currently be trying to store on one computer at this time?

      A whole bunch. One of the difficulties of sequencing the genome is that it's somewhat error prone, which requires that the same region be sequenced several times to make sure that you've gotten it right. The chunks that can be sequenced conveniently are also very small compared to the whole thing, so it's necessary to sequence in overlapping chunks and put it together like a puzzle. The combination of those things means that each base in the genome must be sequenced something like 10 times to get a reliable result. That gets you up to 10 genomes off the bat. Add in the fact that each bit of sequencing information will have meta-data (i.e. where that snippet came from, which machine generated it and when, etc.) associated with it and things fluff up even more. When you start piecing the data together it will require a lot of processing power, so it may wind up being a good idea to use algorithms that trade storage space for processing time, and that can inflate your storage needs even more.

      Even once you have the genome as a finished product, you may very well want to have more than one genome available. An important, but less well publicised, part of the genome project was the decision to sequence the complete genomes of several other organisms at the same time. Those include four species particularly popular among biologists: the mouse, the fruit fly, a round worm used in a lot of research, and baker's yeast. Doing genome to genome comparisons is a very good way of finding the areas that are biologically important because they'll remain similar across organisms. As you can imagine, doing a complete chunk by chunk comparison between two 3 GB data sets can chew up a lot of resources, and having fast access to a huge memory space like that is going to make it a lot easier and faster.

      --

      There's no point in questioning authority if you aren't going to listen to the answers.

    9. Re:New Measurement System? by Anonymous Coward · · Score: 0

      Ha!!!! I can do it in less than 5 min. All 23 chromosomes! Without condom that is.

    10. Re:New Measurement System? by demonbug · · Score: 1
      The field-upgradeable SSD system can hold 27 copies of the Human Genome and transfer data at a rate equivalent to 100 Human Genomes per second.



      let's see, 80 GB/sec divided by 100 Genomes/sec
      uhh, give me a minute here....
      800 MB per genome! Hey, my computer holds WAY more than that! And I thought those genome thingy's were supposed to be complicated!

    11. Re:New Measurement System? by The+trees · · Score: 1

      I'll be impressed by the machine that can turn out clones to keep up with that pace

      --
      $ make work
      make: *** No rule to make target `work'. Stop.
    12. Re:New Measurement System? by Anonymous Coward · · Score: 0

      maybe it's people in the bioinformatics field, the same people who buy crays?

    13. Re:New Measurement System? by p3d0 · · Score: 2
      Anyway, can someone explain this math to me?
      ... a 224-gigabyte Solid State Disk (SSD) with a data transfer rate of 80 gigabytes per second...
      So it could transfer its whole contents in about 3 seconds...
      The field-upgradeable SSD system can hold 27 copies of the Human Genome and transfer data at a rate equivalent to 100 Human Genomes per second.
      So it could transfer its whole contents in about 1/4 second. Which one is right?

      (My guess is they meant "10 Human Genomes per second".)

      --
      Patrick Doyle
      I mod down every jackass who puts his moderation policy in his sig. Oh, wait a sec....
    14. Re:New Measurement System? by radish · · Score: 2

      Phone books, that's what we always used. Or "miles of stacked sheets of typewritten paper". Thats also very useful.

      --

      ---- Den ene knappen er powerknapp, den andre er Bender voice knapp "Bite My Shiny Metal Ass"

  28. Terabytes... by Anonymous Coward · · Score: 0

    Terabytes, I want Terabytes... Say it with me now, Te-Ra-Bytes... No wait!
    Petabytes, I want Petabytes... Say it with me now, Pe-Ta-Bytes... No wait!
    Exabytes, I want Exabytes... Say it with me now, Ex-A-Bytes... No wait!
    Zettabytes, I want Zettabytes... Say it with me now, Zet-Ta-Bytes... No wait!
    Yottabytes, I want Yottabytes... Say it with me now, Yot-Ta-Bytes... No wait!
    Googlebytes, I want Googlebytes... Say it with me now, Goo-gle-Bytes... No wait!
    Uh, what comes after Googlebytes?
    http://www.webstreetstudios.com/scho ol/bitsbytes.h tm

    1. Re:Terabytes... by Anonymous Coward · · Score: 0
      Uh, what comes after Googlebytes?

      My fist.

    2. Re:Terabytes... by RatBastard · · Score: 1

      Googleplexbytes (1 followed by a Google zeros)

      --
      Boobies never hurt anyone. - Sherry Glaser.
    3. Re:Terabytes... by drhemi · · Score: 1
      actually its called a googol

      http://mathworld.wolfram.com/Googol.html

    4. Re:Terabytes... by Anonymous Coward · · Score: 0

      Alright, if you say so, then, here goes...
      Myfistbytes, I want Myfistbytes... Say it with me now, My-fist-Bytes... No wait!
      Uh, what comes after Myfistbytes?

    5. Re:Terabytes... by Anonymous Coward · · Score: 0

      Alright, if you say so, then, here goes...
      Googleplexbytes, I want Googleplexbytes... Say it with me now, Google-plex-Bytes... No wait!
      Uh, what comes after Googleplexbytes?

  29. Running a Samba server... by Sivar · · Score: 1, Offtopic

    To get the full bandwidth utilization out of one of these over a network, you would need at least 640 gigabit ethernet connections, and even that would work only if the cards transfered data at their peak theoretical bandwidth. In reality, you'd likely need a bare minimum of 800.

    That'd be a great headline for Samba.org...

    --
    Computer Science is no more about computers than astronomy is about telescopes. --E. W. Dijkstra
    1. Re:Running a Samba server... by Sivar · · Score: 2

      How, exactly, is this offtopic? It's a storage system, storage systems are used in file servers.

      Sheesh, getting an offtopic moderation for an on-topic post in response to a story that I was the submitter for.

      --
      Computer Science is no more about computers than astronomy is about telescopes. --E. W. Dijkstra
  30. Yeah Yeah... by Loki_1929 · · Score: 2, Insightful

    But can it run Windows fast?

    --
    -- "Government is the great fiction through which everybody endeavors to live at the expense of everybody else."
    1. Re:Yeah Yeah... by Anonymous Coward · · Score: 0

      I'm sorry, but that post is *NOT* insightful.

      Who is modding these posts?! Or is that the new method of trolling? To mod something completely wrong.

  31. Overhead time? by truesaer · · Score: 2, Insightful
    80GB per second is impressive, but the transfer rate of existing drives is also plenty fast. The problem is that setup time, head seeks, and rotational delay make it slow for most data accesses which are small. This is of course the whole point of ram, caching, etc.

    Probably this is just useful for transfers of very large amounts of data, and is the same as other storage devices except for its large size...

    1. Re:Overhead time? by akula1 · · Score: 3, Interesting

      Since this is a solid state storage device its performance will be that of a RAM. This is the main reason why solid state storage is so attractive. There will be no read/write heads etc...

    2. Re:Overhead time? by Anonymous Coward · · Score: 0

      Head Seeks? Rotational Delay?

      Read The Fucking Article. It's SOLID STATE. Fucking incompentant hoser. I hope your boss fires you.

    3. Re:Overhead time? by Dr.+Weird · · Score: 2, Insightful

      Just useful for transfers of very large amounts of data? Gee, is that what a high speed, 224 GB HD is good at? ;-)

      As someone who routinely works with large datasets ( > terrabyte uncompressed) as is typical for physical simulations I would LOVE to have one of these.

    4. Re:Overhead time? by headkase · · Score: 1

      Setup time, head seeks and rotational delay don't apply to this device - it's solid state meaning it's made up of RAM chips.

      --
      Shh.
    5. Re:Overhead time? by Mawbid · · Score: 1

      As someone who routinely works with large datasets, you might want to learn to spell "terabyte". ;-)

      --
      Fuck the system? Nah, you might catch something.
    6. Re:Overhead time? by Dr.+Weird · · Score: 1

      Ouch! No kidding. :-) Unless perhaps I am doing geology... (get it... TERRA-byte. Forget it, I'm an idiot.)

    7. Re:Overhead time? by Anonymous Coward · · Score: 0

      Nah, that was actually rather funny :-)

    8. Re:Overhead time? by stripes · · Score: 2
      Since this is a solid state storage device its performance will be that of a RAM. This is the main reason why solid state storage is so attractive. There will be no read/write heads etc...

      Not all solid state stuff has fast writes. Look at FLASH for example, can be designed to read really fast, but in the general case you have to clear blocks before writing, and the clear is slow.

    9. Re:Overhead time? by Wolfier · · Score: 2

      "Solid State" makes all the
      difference...although I know it is hard nowadays to damage a laptop HDD just by operating it in a car, there were those days a slight knock on a HDD would scratch the disk surface...

      "Solid State" gives you the peace of mind that HDD never will - knowing that you data does not need that wimpy protection from that thin layer of air is good enough - that your data won't be loss all of a sudden if you kick your computer by accident...(although it is hard nowadays, you never know)...

      I know some Solid State storages like CF can only survive so many writes, but that's PREDICTABLE. For a limited time you just KNOW that your data is safe. I can't say that for any HDD, no matter how many G's of shock it can take.

    10. Re:Overhead time? by yomahz · · Score: 2

      The problem is that setup time, head seeks, and rotational delay make it slow for most data accesses which are small.

      The reason why these things exist is because of the mechanical components that make up drives. Solid state storage shouldn't have the same limitation.

      --
      "A mind is a terrible thing to taste."
  32. Slooooowww... by sabre · · Score: 1
    Looks like their web server isn't a Cray...

    -Chris

  33. The Slab by jamesmartinluther · · Score: 1

    I'm convinced that just about every function of the modern computer (save the ui and the cooling system) will be moved into silicon.

    While it will be some time before solid state memory for the home PC becomes practical (and needed), this is but one stop on our journey to a slab computer.

  34. Human Genome by Anonymous Coward · · Score: 0

    Oh yeah? Well, I contain millions of copies of the human genome!! Of course - I can't run very fast...

  35. yeah, pretty impressive by ouslush · · Score: 0

    pretty impressive considering i can't even access their site .....

  36. Even more impressive... by Dynedain · · Score: 1, Redundant

    What makes this even more impressive is that (as the article says) this is solid state storage. Not a magnetic system. Damm I want some!

    --
    I'm out of my mind right now, but feel free to leave a message.....
    1. Re:Even more impressive... by billstr78 · · Score: 1

      There may be commercially available solid state technology in a couple of years, but for now, you have to be a Carnegie Mellon grad to get your hands on one

  37. So, how many by geekoid · · Score: 4, Funny

    Library of congress is that...

    --
    The Kruger Dunning explains most post on /. http://en.wikipedia.org/wiki/Dunning%E2%80%93Kruger_effect
  38. Don't ask.... by rlwhite · · Score: 3, Interesting

    "Nobody who has to ask what a yacht costs has any business owning one." -J.P. Morgan

    Why else do you think a company with expensive products like Cray's would avoid posting prices online?

  39. Teoma should look into this... by Anonymous Coward · · Score: 0

    If Teoma really wants to beat Google, maybe they should look at sinking all their cash into a cluster running these.

    Where as Google's solution apparently has lower throughput than these drives do (they keep most of their web database in ram)

    I'd post links but I can't find any :p

    1. Re:Teoma should look into this... by zeno_2 · · Score: 1

      If you think that Teoma is going to win because its faster your joking yourself. I search for something on google and its returned in about .2 seconds, and its usually right.

      Why would getting that data back to me in .1 seconds make me switch?

  40. cray /.ed? by Anonymous Coward · · Score: 0

    how is that /. never get /.ed?

  41. 80GB transfer rate?!! by Anonymous Coward · · Score: 0

    Shouldn't they have a larger storage capacity that 224GB then... like in the TB range? I mean, jeez with a 80GB transfer rate the disk will fill in 2.8 seconds!

  42. Strage Conicidence by billstr78 · · Score: 1

    Does anyone else wonder if this press release was strangly timed to coordinate nicley with last night's episode of Alias, where they were trying to swipe a cryogenically frozen solid state storage device that was kept 100 feet underground in a terrorist bunker?

    1. Re:Strage Conicidence by colmore · · Score: 2

      right... because Cray is relying on the viewership of *Alias* to boost their sales.

      --
      In Capitalist America, bank robs you!
    2. Re:Strage Conicidence by Anonymous Coward · · Score: 0

      No but the reverse could be true. Don't forget that Cray is owned by SGI who provides alot of the computing power for many of Hollywoods special effects teams. I would be willing to bet that the timing of many network TV episodes is not so random and more of a complex marketing scheme. Remeber that there is big money in getting people do watch a show, so if a little regression analysis is required to time the view of certain media, it would be well worth thier time.

  43. That's nothing.... by Anonymous Coward · · Score: 0

    Apple put a 'supercomputer' into a tiny little cube that anyone could buy. I'm not impressed.

  44. 224 GB / 80 BG/sec = ??? by ksheka · · Score: 1

    Okay. so you can read all of the drive's memory in three seconds. (Not really, since you'll probably never get to the theoretical transfer speed).

    Wouldn't it be more useful if it was 224 TB? If I was going to buy something like this for my PC (!) I would want total storage to be more than three times the bandwidth...

    --
    alias uptime="echo '5:33pm up 22342352324 days, 6:28, 2124315623 users, load average: 2432.40, 12312.31, 123123.19'"
  45. Big Deal by dlb · · Score: 1

    ...and in a year, they'll have doubled the storage and/or speed.

    Solid State is nothing new.
    Storage capacity will keep getting larger.
    Storage access speed will always get faster.

    The day we'll be able to interface it straight into our brains... THAT will be something new.

    Is this really news, besides to the greaseball technophiles who always have to have the latest and greatest the second it gets released?

    ~dlb

    1. Re:Big Deal by rasactive · · Score: 1
      Is this really news, besides to the greaseball technophiles who always have to have the latest and greatest the second it gets released?

      If I were a judgmental person I would judge that you have no fucking idea what you use a Cray for. Technophiles? That kind of technophile is equivalent to a pedophile that goes for sperm during conception. And if you do know a technophile who happens to have purchased one of these fine machines, feel free to give me his phone number and his address. I would love to pay his residence a night-time visit. Solid state is new (at least usable amounts of it). Average storage access speed and capacity have wet dreams about being on the level with these computers.
      That felt good.
    2. Re:Big Deal by dlb · · Score: 1

      It's irrelevant what a Cray is for (or how poorly the company is doing). I never said that any consumer-types could afford it, I said that the only folks who would slobber over this sort of thing are the types of guys who go out and get bigger drives every 4 months.

      Everything will get bigger/faster/smarter... This is nothing new.

  46. Bah. by dkresge · · Score: 2, Informative

    So, is the 80GB/s aggregate access for the (n) processors in the box? It's a Solid State Disk -- In other words, it's memory. And, it's not _that_ fast for a shared memory architecture system. see: STREAM Memory benchmarks

    1. Re:Bah. by Jeffrey+Baker · · Score: 2

      Wow those stream numbers suddenly look really pathetic. On a 32-way DEC^WCompaq EV7, STREAM hauls down 200,000GB/s, or 300 times faster than the top supercomputer in the list you cite.

    2. Re:Bah. by Anonymous Coward · · Score: 0
      What is scary is that an 8 way Sledgehammer system will be able to get aggregate stream benchmarks at over 30GB/s (dual PC2700 memory = 5.4GB/s, *8 = 43.2GB/s, take off overhead, blah). That is comfortably on that table.

      And the processor will probably cost around a grand...

    3. Re:Bah. by Anonymous Coward · · Score: 0
      Bollocks it does. EV7 has 8 RDRAM controllers. Lets assume that run PC1066 memory, or 2GB/s per controller, or 16GB/s per CPU. A 32-way system would get a maximum of 16x32GB/s, or 512GB/s.

      Amazingly enough, this is nowhere near your claimed 200,000GB/s. Maybe you meant 200,000MB/s?

    4. Re:Bah. by Jeffrey+Baker · · Score: 2

      Well, I don't personally make this stuff up. I assumed the huge number is due to cache effects. The 200,000 number comes from presentation given by Compaq people, archived here: http://www.eecs.umich.edu/vlsi_seminar/f01/slides/ bannon.pdf Check page 31. You are right though, after checking http://www.cs.virginia.edu/stream/standard/Bandwid th.html the scale on that chart probably should be divided by 1000.

    5. Re:Bah. by groman · · Score: 1

      I'm not sure exactly, because I don't know much about hammers, but wouldn't PC2700 already account for DDR, so it would be 2700, and shoudlnt' that be Gigabits, not Gigabytes? Feel free to correct me.

  47. How much? by Quixote · · Score: 2

    How much is it, and can I get it as birthday present?

    1. Re:How much? by doooras · · Score: 2

      wow! you have a cray? is it the SV1e or the SV1ex??

  48. LONG way to go.... by Anonymous Coward · · Score: 0

    "The field-upgradeable SSD system can hold 27 copies of the Human Genome and transfer data at a rate equivalent to 100 Human Genomes per second."

    Not very impressive. My testicles hold about 10 million human genomes ... and I.. can transfer ...them ... uhhh... at about 10 million per second.

    Electronics is soooo far behind biology, it's laughable.

    1. Re:LONG way to go.... by Anonymous Coward · · Score: 0

      Maybe, maybe... but what's the packet loss like?

  49. cray.com slashdotted by Vadim+Makarov · · Score: 1
    Their site is very sluggish now, broken graphics, etc.

    Looks like they'll need to optimize it for more traffic, even though they usually don't get much of it. The site is your face. Match its performance to your products, damn.

    --
    17779 eligible voters in a district, 17779 'vote' as one. This is Russia.
  50. Anybody know why it's so much faster? by NanoGator · · Score: 1

    Just curious, I skimmed the press release and couldn't determine how they make it so much faster than existing RAM.

    Is this something Intel could learn, or is it something that just requires tons more money?

    *Concerned that this'll mainly be a big deal for super computers and not end users.*

    --
    "Derp de derp."
  51. Cray's site... by kypper · · Score: 1, Offtopic

    According to netcraft,
    "The site www.cray.com is running Apache/1.3.22 (Unix) mod_ssl/2.8.5 OpenSSL/0.9.6b on Solaris 8."

    That has to be one crappy connection to divebomb solaris.

    1. Re:Cray's site... by mrm677 · · Score: 1, Offtopic

      That has to be one crappy connection to divebomb solaris.

      What are you talking about? I've got Solaris 8 running on my Ultra 5 which is about as fast as a 180MHz Pentium Pro.

      Surely because its running Solaris 8, the connection must suck then??? I don't think so.

    2. Re:Cray's site... by Anonymous Coward · · Score: 0

      No solaris actually equals magic, and it will turn any machine into fort knox just by installing it. I guess you didn't get that last update =)

    3. Re:Cray's site... by groman · · Score: 1

      More importantly, why would Cray be running Solaris 8? Isn't Cray at least partially owned by SGI? Isn't SGI pushing their own servers with IRIX?

      feel free to correct me if I'm wrong, but this feels like a mini-"We Have The Way Out" deal all over again.

    4. Re:Cray's site... by Anonymous Coward · · Score: 0


      The Cray business unit was purchased from SGI by Tera Computer. The merged company is called Cray Inc.

      http://www.cray.com/news/0003/acquisition.html

  52. i doubt it is enything revolutionary by Edmund+Blackadder · · Score: 2

    i bet they just use more parralel data buses, hard drives, whatever.

  53. Transferring my genome by Joel+Ironstone · · Score: 5, Funny

    I can transmit my whole genome in a few seconds, While the silicone guys find ways of speeding this benchmark up, I'm looking for ways of slowing it down.

    1. Re:Transferring my genome by nytes · · Score: 1

      I can transmit my whole genome in a few seconds, While the silicone guys find ways of speeding this benchmark up, I'm looking for ways of slowing it down.

      Maybe you're bit-banging too fast?

      --
      -- I have monkeys in my pants.
    2. Re:Transferring my genome by Aceticon · · Score: 5, Funny

      The "natural" way of transmiting the human genome (at least for half the human beings) is high speed but also high latency - it takes a while to start it up (how much depends on the individual) but when it starts it's a burst of genomic information.

      Strangely enough it's all contained on packets with a size of 1/2 human genome...

    3. Re:Transferring my genome by khuber · · Score: 1

      Where's the mod option for "eye roller"?

      -Kevin

    4. Re:Transferring my genome by Anonymous Coward · · Score: 0
      I can transmit my whole genome in a few seconds

      I don't think transferring it to a wad of kleenex counts.

    5. Re:Transferring my genome by Jah-Wren+Ryel · · Score: 1

      Don't underestimate the bandwidth of a VW Van filled with refrigerated test tubes.

      --
      When information is power, privacy is freedom.
    6. Re:Transferring my genome by Geekboy(Wizard) · · Score: 1

      yay, goatse.cx re-direct. I was waiting for one of those.

    7. Re:Transferring my genome by bonzoesc · · Score: 1

      Just burning off some karma for rankings.

    8. Re:Transferring my genome by JunkDNA · · Score: 1
      I can transmit my whole genome in a few seconds

      You have a point. You've got a pretty amazing transfer rate. Especially when you consider ~180 million nearly identical copies of your genome (at ~8.3 gigs a pop) get transferred in a few seconds, it makes for one hell of a RAID system!

    9. Re:Transferring my genome by BlueUnderwear · · Score: 2

      For an extra laugh, read the last line of this page. Apparently, BonzoESC didn't heed that advice...

      --
      Say no to software patents.
    10. Re:Transferring my genome by Bruha · · Score: 0, Redundant

      *dies*

      I'm sorry 5 was a understatement for this being funny.

      Best laugh I've had all day. Back to crying about being laid off by Verizon (backstab) Wireless.

      Need an network engineer?

    11. Re:Transferring my genome by PaulBu · · Score: 1

      Classical quote from parallel computing lore
      sounds quite appropriate here...

      A woman can make you a child in 9 months,
      but 9 women would not be able to do that in a month...

      Paul B.

    12. Re:Transferring my genome by Anonymous Coward · · Score: 0

      No but once the data comes back... 9 babies in 9 months comes out to 1 per month. Stats are great.

    13. Re:Transferring my genome by pacc · · Score: 2

      I can transmit my whole genome in a few seconds
      I don't think transferring it to a wad of kleenex counts.

      And wouldn't that be premature anyway.

    14. Re:Transferring my genome by atamar · · Score: 1

      > Strangely enough it's all contained on packets with a size of 1/2 human genome...

      It's called lossy compression, and they do all sorts of optimizations to achieve that.
      The quality of the duplicate is never comparable to the original.
      Of course, a lot depends on the codec of the reproduction implementation; some produce better results than others. You probably want to read through tests (try Google) before deciding on the one you want.

  54. ummm. . . no by Bastian · · Score: 5, Informative

    The CPU gets stuff from the cache.

    The cache gets stuff from the RAM.

    The RAM gets stuff from the hard drive.

    The solid state machine won't act like faster memory, making cache misses cost less. It will act like a faster hard drive, making page faults cost less. Using this stuff as a substitute for RAM will slow down your computer unless you have it hard-wired into your system's bus in place of RAM.

    1. Re:ummm. . . no by Anonymous Coward · · Score: 0

      Which, sense this stuff is probably just SDRAM, would be kinda like well, ram.

    2. Re:ummm. . . no by Anonymous Coward · · Score: 0

      Since the transfer rate is 640 Gbps which is 40 times faster than SDRAM, it's probably not SDRAM.

    3. Re:ummm. . . no by Anonymous Coward · · Score: 0

      Actually, I believe this is exactly what Cray intends to do. Replace main memory with this SSD.

    4. Re:ummm. . . no by cpeterso · · Score: 1


      no wonder their web site is slashdotted. ;-)

    5. Re:ummm. . . no by DarrinWest · · Score: 1

      So just mmap it as a device. It'd go through the IO bus and skip RAM, no?

    6. Re:ummm. . . no by stripes · · Score: 3, Informative
      Since the transfer rate is 640 Gbps which is 40 times faster than SDRAM, it's probably not SDRAM

      It has 40x the transfer rate, but they don't talk about latency, so maybe it is "just" SDRAM, but 1280 bits wide not 32 (or 1440 bits with some sort of ECC).

    7. Re:ummm. . . no by Omnifarious · · Score: 2

      No, it wouldn't. It just wouldn't be copied from kernel memory to user space memory.

      Also, the PCI bus is lots slower than the memory bus, by a factor of 4 or 8.

    8. Re:ummm. . . no by pixel.jonah · · Score: 1

      You think they have a PCI card to connect this to your computer?!?!? (Maybe it's a 64bit 66mhz one, no?)

    9. Re:ummm. . . no by Anonymous Coward · · Score: 0

      > The CPU gets stuff from the cache.
      >
      > The cache gets stuff from the RAM.

      Actually, most Cray CPUs don't have a data cache. Data caches tend to get very low hit rates on scientific applications which mostly just loop through huge arrays: in this case the cache just gets in the way by adding a extra latency.

  55. Solid State Disk? by Anonymous Coward · · Score: 0

    They call it a "Solid State Disk (SSD)"? Sounds like a contradiction in terms to me. Usually "solid state" means "on chip(s)", and a disk is, well, round. Can anyone confirm/deny that there's a disk inside? Looks like the marketing department tried to justify its "usefulness" by changing a word along the way. The submitter calls it "storage" which looks more like the right term.

    Or maybe they *do* have a spinning disk inside, and they write "solid state disk" to mean "non-liquid disk"...

    1. Re:Solid State Disk? by TheBigDinK · · Score: 1

      Well if you want to get really picky, I believe it is a "disc" that is round, as in "compact disc" rather than "compact disk."

      A "disk" on the other hand, seems to be more of a generic storage term. Even HDs are referred to as "disk" because the round platters are just part of the insides, we don't see them.

    2. Re:Solid State Disk? by Anonymous Coward · · Score: 0

      I believe it is a "disc" that is round

      Nope. Every math book call the set {(x,y) | x^2+y^2 <= 1} a disk, written with a "k". "Disc" is american spelling reserved for optical media.

    3. Re:Solid State Disk? by Anonymous Coward · · Score: 0

      >Nope. Every math book call the set {(x,y) | x^2+y^2 <= 1} a disk, written with a "k". "Disc" is american spelling reserved for optical media.

      Well, my OED minidictionary (which would be the non-American spelling / definition source) says this about "Disc":

      disc n. thin circular plate or layer; thing shaped thus; record bearing recorded sound; non-magnetic circular storage-device for data; magnetic disk (see magnetic).

      and:

      disk n. = disc; = magnetic disk (see magnetic).

      So, double nope to that. A disk is a disc is a record is circular is magnetic (etc) across the entire planet (unless you want to argue that we Canadians define disc differently from Americans or the British).

    4. Re:Solid State Disk? by emmons · · Score: 1

      Duh, the chips are round.

      --
      Do you even know anything about perl? -- AC Replying to Tom Christiansen post.
    5. Re:Solid State Disk? by Anonymous Coward · · Score: 0
      This clearly illustrates why you shouldn't trust a minidictionary as an authorative source.

      (I won't even go into the fact that it's Canadian.)

  56. The skinny on the SV1 SSD by Anonymous Coward · · Score: 1, Informative
    So, is the 80GB/s aggregate access for the (n) processors in the box? It's a Solid State Disk -- In other words, it's memory. And, it's not _that_ fast for a shared memory architecture system.

    Actually, if I recall the details correctly, this only exists because the SV1 is a 32-bit system.

    "But it's a 64-bit processor!" I hear you exclaim. Yes, but PVP Crays (that includes EL, C90, J90, and SV1, but not T90 or T3D or T3E or SV2) have only (up to) 32 bits of physical memory address.

    "But it supports 64 GB of memory, 32 bits is only 4 GB!" True, if you count in bytes! The Cray PVP line addresses memory in 8-byte words, so 2**32 words is 64 GB.

    Thus, the engineers had a dilemma. They could fit 256 GB into the box (due to denser SDRAM parts), but the CPUs could only address 64 GB of it directly!

    So they stuck a smart memory controller in the box, and did some OS hackery, and voila! the extra memory became available as a (very fast) disk instead. Additionally, the memory controller itself has a builtin memcpy engine, so if you're just copying data from "disk" to "ram" it doesn't touch the CPU at all, the CPU just tells the controller "hey, copy me 500 pages from address A to address B" and the controller does it.

    I don't know offhand if the 80 GB/sec is the CPU-to-SSD bandwidth, or if it's the peak bandwidth of the copy engine...

  57. Check my math, but... by cmckay · · Score: 1
    I think their performance analogy about the human genome is a bit off. From the article:
    1. Their 224-gigabyte Solid State Disk (SSD) can hold 27 copies of the Human Genome
    2. The SSD can transfer data at 80 gigabytes per second, a rate equivalent to 100 Human Genomes per second.
    #1 yields a Human Genome size of 8.3GB (224GB holds 27 genomes), but for #2 to be true, the genome would have to be 800MB (80GB/s == 100 genomes/s).

    Did they bork their numbers, or am I just misunderstanding?
    1. Re:Check my math, but... by glwtta · · Score: 2

      That's the beauty of the new HGPS (Human Genome Per Second) unit of measure - there isn't actually a "size" for the genome, since the "genome" as such is a very vague concept, when it comes to actual computer data. So if it's anywhere between 100MB and 500GB it still qualifies.

      --
      sic transit gloria mundi
    2. Re:Check my math, but... by Anonymous Coward · · Score: 0

      Think you are right... Press release written was not only written by non-techies as ususal but also be people without any math skills...

    3. Re:Check my math, but... by Anonymous Coward · · Score: 0

      http://www.pbs.org/wgbh/nova/genome/facts.html
      "T he entire human genome requires three gigabytes of computer data storage space. (One million base pairs of sequence data equals one megabyte of storage space; the human genome has three billion base pairs.)"

  58. Playstation 2 (aka PS2) by Jagasian · · Score: 1, Troll

    Well, it doesn't beat my Playstation 2! It's 100 times for powerful than a super computer!

  59. That's not that much by pmarley42 · · Score: 1

    224GB isn't that much. The only thing that's amazing is the 80GB/s

    1. Re:That's not that much by Sivar · · Score: 2

      224GB is quite a bit for a solid state storage device--look at the Quantum Rushmore drives. Their capacities are in the single digit GB.

      --
      Computer Science is no more about computers than astronomy is about telescopes. --E. W. Dijkstra
    2. Re:That's not that much by 2cool4school · · Score: 1

      Hang on a minute, is 80GB/sec any use without RAM that is at least as fast? Wouldn't there just be a huge bottleneck? Or is it the case that there is some technology that makes this irrelevant? Or is the drive put to some use that the RAM speed doesn't matter? Is the RAM in CRAY's computers faster than 80GB/sec? Or is this storage a product of 'blue sky' research and doesn't have an application yet? Sorry about all the questions, thinking into the keyboard! Very impressive though!

  60. Marketroid sense is tingling by Bastian · · Score: 5, Interesting

    ncludes a 224-gigabyte Solid State Disk (SSD) with a data transfer rate of 80 gigabytes per
    second


    can hold 27 copies of the Human Genome and transfer data at a rate equivalent to 100
    Human Genomes per second


    Ok, so can it hold more data than it can transfer in a second, or can it transfer more data in a second than it can hold? Pick one, boys.

    1. Re:Marketroid sense is tingling by lkaos · · Score: 2

      Obvious typo. Probably meant to say it could hold 270 copies of the Human Genome. Although I seem to remember the Human Genome being slightly biggerthan the .8GB they are claiming it to be.

      *me looks aside to the 5 640MB cd's labeled 'Human Genome'.

      --
      int func(int a);
      func((b += 3, b));
    2. Re:Marketroid sense is tingling by Anonymous Coward · · Score: 0

      Wow, you have the human genome on CD??? Where did you get that?!

    3. Re:Marketroid sense is tingling by Anonymous Coward · · Score: 0

      He put his finger in the CD burner...

    4. Re:Marketroid sense is tingling by jigma · · Score: 1

      More likely meant to be 10 Genomes a second

      --
      "linux is only free if your time has no value" - Jamie Zawinski
    5. Re:Marketroid sense is tingling by lkaos · · Score: 3, Interesting

      Actually, I think they were fibbing and not mentioning that they meant compressed Genomes, in which case, the numbers work out perfect if they meant 270 and 100.

      --
      int func(int a);
      func((b += 3, b));
    6. Re:Marketroid sense is tingling by Anonymous Coward · · Score: 0

      Please tell us where you got them? I'd really like to know. Sure, I would have no real use for them, but it would just be a great part of a digital collection, don't you think?

    7. Re:Marketroid sense is tingling by Anonymous Coward · · Score: 0

      Why, from humans I'd imagine. I'd be afwul suspicious of anyone having claimed to have gotten the human genome from pigs or goats.

    8. Re:Marketroid sense is tingling by lkaos · · Score: 2
      --
      int func(int a);
      func((b += 3, b));
    9. Re:Marketroid sense is tingling by Anonymous Coward · · Score: 0

      Oh wow! Thank you very much :-)

  61. Power requirements in JDU? by UsonianAutomatic · · Score: 2, Funny

    Ok, if we're going to measure capacity in terms of Human Genomes, I want to know how many Jelly Donut Units per hour it takes to power this thing.

    "From the intensity of the flame we can deduce that this was a particularly delicious donut."

  62. Ohmygod... by tandr · · Score: 1

    This one is serious music jukebox...

  63. Won't somebody think about what this really means? by Anonymous Coward · · Score: 0

    That's 80Gb/s of streaming porn!!!

    All you need is the fat pipe to match.

  64. well. damn. by doooras · · Score: 2

    I've looked all through the Cray website, and I can't find the online order form. How am I going to get one of these systems FedEx'd to me now?

    1. Re:well. damn. by Jouster · · Score: 2, Interesting
      Actually, you have to email ussales@cray.com. Here's a recent email I received from them regarding prices:
      Date: Fri, 29 Mar 2002 12:51:56 -0600
      From: Cathy Wells
      Subject: Re: Prices
      To: Dan Reif
      Organization: Cray Inc.
      X-Mailer: Mozilla 4.79 [en] (Windows NT 5.0; U)
      X-Cray-VirusStatus: clean

      The Cray SV-1 vector supercomputer is the current production product for vector types of application problems and most legacy Cray scientific and engineering applications. A 32 processor SV-1 with 32 megabytes of shared main memory would have a list price over $3M depending on the configuration and peripherals required.

      The Cray T3E massively parallel supercomputer continues to be the world record holder for the fastest sustained performance on a real world application. These systems can scale to over 2000 processors. An entry level configuration for the Cray T3E might start at approximately $2M with a 1000 processor system listing at over $30M depending on the configuration.

      The MTA-2 will be available 3Q 2001. It will be an all CMOS machine--much simpler to manufacture than the all GaS system at SDSC. The MTA-2 will have between 16 - 256 processors and 64 - 1024 GBytes of shared memory. We plan to build the MTA-2 in 16P cages. A case, or stack as we're calling it, will house 4 cages. The case will be approximately 3 ft wide, 5 ft deep, and 7 ft tall. It will weigh a few thousand pounds.
      Jouster
  65. just for comparison by adminispheroid · · Score: 1
    From the article:
    The field-upgradeable SSD system can hold 27 copies of the Human Genome and transfer data at a rate equivalent to 100 Human Genomes per second.
    And all the screwing in the world only transfers about 2 human genomes per second.

    (If you just count the ones that, you know, like, make it.)

    1. Re:just for comparison by user32.ExitWindowsEx · · Score: 1

      If the transfer rate is 80 GB/sec and the capacity is 224 GB, how can 224 GB hold only 27 copies of the human genome while 80 GB/sec be enough speed to relay the same data 100 times a sec?
      I figure someone got the numbers backwards. They meant to say that the thing holds 100 copies of the human genome and relays 27 copies per second.

      --
      "Evil will always triumph because good is dumb." -- Dark Helmet
  66. wtf? by bo0push3r · · Score: 1

    for the announcement of something so impressive you'd think they'd have at least had someone proof-read the press release. it looks like my 8-year-old nephew approved the final copy.

    1. Re:wtf? by Anonymous Coward · · Score: 0

      I'll have you know that I am a highly competent proof-reader.

      - Your 8-year-old nephew.

  67. Actually... by Chazmati · · Score: 4, Funny

    The siliCONE guys might help to shave a couple seconds off your time. It's the siliCON guys who do the computer thing.

    1. Re:Actually... by Jonny+290 · · Score: 1

      Right.

      I can see it now. Slashdotters bitching because their "lite" version holds only 100 gb, and transfers at only 20 gb/sec, costing anything more than 300 bucks.

      This has absolutely no relevance to the desktop PC world unless you're NFS mounting this fucker over an OC-192 plugged straight into your network. And even then a good fast SCSI chain will do just as well.

      --
      Hey Taco! Looks like you're using the "infinite monkeys and typewriters" scheme to generate Ask Slashdots again...
    2. Re:Actually... by pmsr · · Score: 1

      ... and mod and tweak sites drooling because they found a way to pump up the speed by changing two SMD resistors in the main circuit board. :-)

      /Pedro

  68. Re:Benchmarks by Brendor · · Score: 1

    I don't know about Quake III, but I do know This Is The Ultimate storage solution for my ASCII Art colection.

  69. Mom says dinner is ready. by Anonymous Coward · · Score: 0

    Come get it before it's cold.

  70. Why use the humane genome as a reference? by mbessey · · Score: 2

    Because that's who they're marketing it to - the bioinformatics crowd. Try to forget that there is little-to-no evidence that there's actually a market for all this genome-crunching that's going on.

    Or that there are real questions about the quality of the sequencing that's been done, not to mention the (abysmal) quality of the code being written to analyze the sequences.

    Bioinformatics is the dot-com boom all over again...

    -Mark

  71. Hello McFly by ToasterTester · · Score: 1

    "includes a 224-gigabyte Solid State Disk (SSD)"

    The author needs to look up what a solid state disk drive is. A solid state drive in simple terms is memory in the shape of a disk drive. So that's why it's as fast as memory. Now for fun get your calculator out and figure out what 224 GB of RAM cost.

  72. What about RAM??? by tweakt · · Score: 2
    If they can make a storage system that moves 80GB/sec, why are we still putzin' along with DDR RAM that only moves 4.2GB/sec ???

    Methinks were being jipped...

    1. Re:What about RAM??? by Spinality · · Score: 1

      Methinks they charge a little more for their hardware than we're paying for our RAM.

      --
      -- We all have enough strength to endure the misfortunes of other people. La Rochefoucauld
    2. Re:What about RAM??? by Anonymous Coward · · Score: 0

      i dunno man, have you checked the prices lately?

    3. Re:What about RAM??? by guile*fr · · Score: 1

      but does your RAM survive a "pull-the-plug"? it's SSD you know...

  73. core memory and persistent operating systems by dmoen · · Score: 3, Interesting
    Back in the 60's, non-volatile magnetic core memory was used instead of RAM. So the idea of non-volatile memory is actually very old.

    With virtual memory hardware, you can write an operating system that simulates non-volatile main memory, using hard disk as a backing store. What you get is a Persistent Operating System. You don't need a file system. Instead, you store data structures in main memory, and they persist forever, surviving reboots.

    Doug Moen.

    --
    I have written a truly remarkable program which this sig is too small to contain.
    1. Re:core memory and persistent operating systems by klui · · Score: 1

      Definitely needs a new name/acronym if it has any bugs...

    2. Re:core memory and persistent operating systems by claeswi · · Score: 1

      I've recently worked some on hacking PalmOS C, and one of the first things you have to get used to is that everything resides in its (solid-state) RAM, which is divided into different heaps for different things (dynamic, runtime, etc.) but where programs are _not_ moved from the static heap to dynamic RAM at runtime but are rather run in-place (analogous to running directly from the hard drive in a desktop computer), which is pretty strange.

      --
      I'd like to believe that when the right woman comes along I'll have the courage to say, "no thanks, I'm married."
    3. Re:core memory and persistent operating systems by javilon · · Score: 2

      The problem with one of those is that you can't reboot :-)

      --


      When his defense asked, "Which computer has Jon Johansen trespassed upon?" the answer was: "His own."
  74. Wow, how terribly inefficient. by mbessey · · Score: 2

    Each base pair only needs 2 bits to represent it (there are only 4 bases). Based on the genome-munging code I've seen, though, it's very common to represent genome sequences with one byte per base pair, like so:
    "CGAAGAACGAT"

    A little comp. sci. 101 would be a good investment for some of these people, I think.

    -Mark

  75. CERN's LHC by sisa · · Score: 1

    Only projects that easily comes to my mind for using such a technology is CERN's LHC project. Though it seem that this one is still inadequate for their purposes. Promising development in any case. Some numerical information on LHC computational challence.

  76. Actually, only half... by berck · · Score: 1

    Actually, you're only transmitting half your genome....

    1. Re:Actually, only half... by Anonymous Coward · · Score: 0

      nah he's transferring all of it. Just not all of it is used.

  77. Bottleneck by kyoko21 · · Score: 1

    Interesting enough, even if Cray's webserver uses these fast disks, their uplink probablly can't do 80GBps transfer....well unless they have some crazy darkfibers that no one uses.... :-)

    Would be cool though, to have 80GBps uplinks to download all the cool media you want.... whew! Let see....224/80 ... that would mean it would only take less roughly 3 seconds to fill up this drive.... wow.... :-)

  78. Human genome transfer by wringles · · Score: 3, Funny

    "System Can Transfer 100 Copies of Human Genome Per Second"

    Big deal, I can transfer O( 1e6 ) half-copies of the human genome in less than five minutes.

    1. Re:Human genome transfer by Anonymous Coward · · Score: 0

      LOL

    2. Re:Human genome transfer by quintessent · · Score: 2

      Heh, my car can transfer a few trillion in a lot less than that.

  79. Tech details by PhotoGuy · · Score: 2

    The announcement was pretty thin on technical details. What exactly is meant by "Solid State Disk." Are there spinning platters? That title implies not, to me. Exactly what technology allows 224G of storage in non-platter form? Is this an actual commercial implementation of the crystal holography gunk and other amazing "future" stuff?

    It sure would be interesting to know if this is a real advance, or just a big disk.

    -me

    --
    Love many, trust a few, do harm to none.
    1. Re:Tech details by tcc · · Score: 3, Informative

      >What exactly is meant by "Solid State Disk." Are there spinning platters?

      No moving parts, you can look at this like a "big ram disk" exept it has it's interface like another storage device. Look at this like A compactflash for example (it's not "SSD" but it's a good comparison.

      There are a lot of interfaces (PCI, ATA, SCSI, proprietary (80GB/sec is either a big aggregated pile of raids or something similar) for these "drives" at various price points. The advantage of a SSD drive on a PC is that you have instant access, and it moves the stuff at a lightning speed limited only by your bus. Let's say you run a SSD drive on a Ultra160 interface, what you'd probably see with a disk benchmarking tool is 100nS access time (versus ~10ms for a standard drive) and you could see the real-world numbers of your scsi bus, probably around 140MB/sec (didn't try one on a U160 bus). The application for these babies are numerous: instant access to data on boards that don't handle 100GB of ram to cache everything or you wanting the machine to preload 1 hour at every reboot, bandwidth hungry application (although a raid could do the same here, but I saw some specific application needed both the bandwidth and under 1ms access time needed so..), for heavy swapping of numbers without using a buttload of ram again, etc.. probably some other people could think of something other. Usually when you break a certain amount of GB, the drives becomes cheaper than a motherboard that could handle a load of ram and the ram modules themselves, so it makes more sense if $$ is a factor (but still it's very expensive, we're talking 10K+ easily for a few GB).

      There's also plenty of product on the net (search google), like I said, some are PCI cards that you add to your system with PC100 ram on it, some are IDE/SCSI, etc.. But for home people, you'd be better off with a cheap IDE raid card and a few drives, it's way cheaper :).

      --
      --- Metamoderating abusive downgraders since my 300th post.
    2. Re:Tech details by yzquxnet · · Score: 1

      I would guess that a solid state design would be far more durable as well. I know I've abused other SSD chips and had them take the abuse. Spinning HD's are really easy to damage. One signifigant jostles and you can crash the head.

      What I'm amazed at is just how big of a jump it really is. Even in terms of supercomputing.

  80. It seems to me by compupc1 · · Score: 4, Informative

    It seems to me that many people here have no idea what a true supercomputer (or more specifically, a Cray) is or what makes it different from a simple cluster. Here's a few things to think about:

    - Crays do not have monitors. They do not have keyboards, or mice.
    - Crays do not run Windows. Crays do not run Linux. Crays usually run UNICOS, a special *nix designed specifically for Crays.
    - Crays communicate with the outside world through a host terminal, like a SGI workstation, or something similar to that. Crays DON'T HAVE CD-ROM DRIVES!
    - Nobody but those with 8-9 figure incomes get to buy a Cray. They cost MILLIONS, and the higher end ones can cost many many tens of millions.
    - Pretty much the type of people that WOULD buy a Cray would be the government, and very very large corporations. Sorry, guys.
    - Simply connecting 30 PCs together in a cluster will result in a nice, fast supercluster, but it won't come close to a Cray, because Crays are designed from the beginning to be as parallel as possible. Face it: beowulf clusters really can't make the best use of the contained hardware because the hardware wasn't designed to be so distributed.
    - Be impressed with Crays. Be very impressed.

    --
    -James
    1. Re:It seems to me by Tazzy531 · · Score: 2
      Pretty much the type of people that WOULD buy a Cray would be the government, and very very large corporations. Sorry, guys.

      Even then, Cray cannot sell them to many countries because of legal restrictions... :-)
      --


      _______________________________
      "I'm not Conceited...I'm just a realist..."
    2. Re:It seems to me by cyr · · Score: 1

      Old Crays also make cool sofas, which adds a lot of value to their multi-million dollar price tags. =)

      We have two of them (a Cray-1 and a X/MP) here at LiU

    3. Re:It seems to me by akhaksho · · Score: 1

      So are you saying that a cluster isn't a supercomputer? Check out the Top 500 list. Notice how many Crays are in the top 15? ASCI White is a cluster and is definitely a supercomputer. Clusters are the future of supercomputing.

    4. Re:It seems to me by mikael_j · · Score: 1

      yes, but the other clusters aren't exactly run-of-the-mill Linux clusters put strung together with shoestrings, they are for example Alpha clusters running OpenVMS or Tru64...

      /Mikael

      --
      Greylisting is to SMTP as NAT is to IPv4
  81. Great :-/ by KewLinux · · Score: 1

    I've got the pr0n, no doubt about that. But how in the world am I going to get (or even pay for) a connection that will support 80GB/s downloads now that I have the storage....

    --
    fear my zig!
  82. -OT- usefulness of bioinformatics by jonbrewer · · Score: 4, Interesting

    (to diverge ever so slightly)

    "Bioinformatics is the dot-com boom all over again..."

    I think not.

    There is quite a market for bioinformatics. My employer spends around 5 billion USD a year on pharma R+D. Much of that money is used in traditional "brute-force" type attacks of screening many compounds against many targets.

    There is tremendous potential for savings through bioinformatics, and the evidence is working its way through pharma pipelines as we speak.

    While there may be as much hype around bioinformatics, the field is solving a genuine problem for a mature, well-funded industry, unlike the dot-com book which speclated on products many didn't want with money that didn't exist.

  83. 80 GB/s transfer speed. by IdleTime · · Score: 1

    I'm sorry, but this baby is not good enough!

    80 GB/s and only 224GB total storage= 224/80=2.8 sec to fill the whole disk. I.e worthless! What am I going to do after the first 2.8 seconds of usage????

    Stupid Cray! LOL!

    --
    If you mod me down, I *will* introduce you to my sister!
    1. Re:80 GB/s transfer speed. by Ilgaz · · Score: 1

      Its used as RAM, not as DISK.

      Wondering when first post appears claiming his Intel/AMD system has more GB's... :-)

  84. Because they need it ! by Anonymous Coward · · Score: 2, Insightful
    Read the press release:

    "With their 32-gigabyte central memories..."

    Of course they need a 224 GB "solid state device" ! Every worthwile competitor of theirs can just put 256GB of main memory in their big box.

    It looks to me that Cray can't easily address more than 32GB on their box, so they just use "extended memory" as a disk.

    Buy an IBM / HP / Sun top of the line, stack
    it with 256GB, and you can use 224GB as a file buffer. Or 128GB, or 16 GB, or whatever you do not use for something more important.

    You've been fooled by PR spin on a limitation :-)

    Like windows and 36bit addressing on Xeons...

    1. Re:Because they need it ! by angio · · Score: 5, Informative

      Not really - consider SGI's servers, for instance. The Origin 3800 can handle 1 TB of RAM -- but it's a CC-NUMA machine, meaning you have to go through an intermediate router (don't think Internet; much faster) to get to the memory. SGI machines have a limit of 8GB per processor "brick", and their bricks interconnect at 1.6 or 1.2GB/s.

      Then consider the SunFire 15K - it's an SMP machine; processors fit on boards that can contain up to 32GB of RAM; after that, you have to go off-board through a switch to get to other memory. Each system board has about 9.6GB/s of offboard memory access speed.

      In short, Cray isn't tooting needlessly - this is impressive bandwidth to the memory. Latency is probably fairly high on it, but for streaming vast quantities of data in and out of local storage, it's probably amazingly nice.

    2. Re:Because they need it ! by Anonymous Coward · · Score: 0

      I am quite familiar with high-end machines design, thanks.

      So you so believe that Cray's 80GB/s figure is from one processor, or even from one node ? Get real.

      Until proven otherwise, I'll presume that 80GB/s is the total simultaneous bandwith available for concurrent accesses from each processor, and to local or locally cached data.

      That is still very respectable, but not earth-shattering or as drool-worthy as some posters may think.

  85. Connecting with the E-Bomb... by Devil's+BSD · · Score: 0

    First off, it's nice to see Cray coming back into the spotlight again. But on to my main subject: Imagine if they hooked a couple of these Cray storage racks up to the ASCI White system at Los Alamos and did some calculations. I wonder how many teraflops that would give it - I mean, 80 gigs a second is probably more memory bandwidth than 1 board of ASCI White could handle. Who knows, 50, 60 teraflops could easily be within reach. Problem arises though, how much cable would you need to provide the necessary bandwidth to the system?!

    --
    I'm the Devil the Windows users warned you about.
  86. I don't know about you... by evilpaul13 · · Score: 2

    But I'm picking up to after work!

    /dies

    I'm not sure what happened, but my coworker just screamed "the price" and died in his cubicle?

  87. Actually... by redcliffe · · Score: 1

    if Cray wants to cement it's position in computing well into the future, they should transfer this technology to high-end PC's. There would be a huge market for it at the right price...

  88. anyone have the rice link? by Anonymous Coward · · Score: 0

    the link where it says that the rice genome has more information in it than the human one?

  89. A bit like the Sydharb by BrokenHalo · · Score: 1

    I'm serious - this is the rather evocative name of a unit of capacity equivalent to that of Sydney Harbour - can't remember whether it's at high or low tide... :-)

  90. Google uses solid state by niola · · Score: 2

    We had this discussion on slashdot back in the day and it seems to be a trend. The fact is that with all the progress made in computing technology, those spinning platters and movable arms have been the bottleneck for some time.

    --Jon

    1. Re:Google uses solid state by Nobody's+Hero · · Score: 0

      It's so very true...
      What I'm curious about is when something like this will become practical for an homw user. I can only imagine the stacks of cash required to purchase this. Not to mention the cost of the equipment to utilize it fully.

      --
      The Only Person Willing to be Me is ME!
  91. Ho hum. by sommerfeld · · Score: 1

    Six months ago, sun announced starcat which can take 576GB of main memory. Sun also claims 172.8GByte/sec bandwidth across the interconnect.

  92. Is that a new unit of measure I see? by Anonymous Coward · · Score: 0

    "100 Human Genomes per second" or 100 hg/s. Someone needs to submit "hg/s" to the ISO people as a proposal for a high speed data transfer unit. Of course, they'll have to decide whose genome to use as a benchmark, and then micro-engrave it on platinum plates, but hey they love that kind of thing.

  93. further stats released.... by panck · · Score: 1

    Follow the link, and you'll see some even more impressive benchmarks:

    Install cost: $200T
    this is equivalent to 150 United Kingdom GDP's!
    Maintainance cost: $40B/ms
    or on average $40T per second of cash straight into Cray's pocket!
    This is equivalent to the amount of money 240 Bill Gates' lose when he stops to pick his nose!

    --
    "What thou shalt not, I shalt did!" -Bart Simpson
  94. What's the cost per byte? by Animats · · Score: 2
    This is basically lots of DRAM in a box, right? Not a technical breakthrough.

    A few racks of 1U servers could be configured to have that much DRAM.

    1. Re:What's the cost per byte? by Tazzy531 · · Score: 2

      Difference is..when someone trips on the power cable, this one doesn't lose all of its data. Whereas try doing that in a computer lab during finals week and see if your data in DRAM is gone. I think I remember seeing that Google was using stuff like this for their storage rather than HDD.

      --


      _______________________________
      "I'm not Conceited...I'm just a realist..."
  95. tingling ? by Erris · · Score: 3, Funny
    Marketroid: Clever observation. Of course it all depends on which Genome we are talking about really. Would that be an African or a European swallow?

    Bastian: I don't knooooowwWWWWWW, AAAHHhhhhhh! Bastian is trown from the clif by an an invisible hand.

    Those struck by lightening and survive fear tingling sensations.

    --
    DMCA, Hollings, Palladium. What might have sounded like paranoia is now common sense.
  96. Memory/RAM by sean23007 · · Score: 3, Interesting

    So can we expect them to design a new type of system that has non-volatile memory and vast storage in a similar array, divvied up on the fly by the system depending on whether it needs storage or memory at the moment? I've been waiting for the day when memory and hard drive became one, and this seems to take that one step closer to the inevitable.

    --

    Lack of eloquence does not denote lack of intelligence, though they often coincide.
  97. corporate whoring by Anonymous Coward · · Score: 0

    If I just wanted a website that spit up corporate press releases, I would go to Yahoo finance. There are a lot of interesting stories we could have about Cray, but this isn't one of them.

  98. WOW indeed by samantha · · Score: 2

    What the heck kind of bus do the expect to drop this wad of data onto? Or are they planning to just map it directly into some memory array? Something like this could change a lot of software - having offline storage faster than main memory is a big deal for many algorithms. The implications are huge! When can I get one and how many lotteries do I need to win?

  99. Speedy Renderings? by Anonymous Coward · · Score: 0

    I'm not much of a hardware buff, but wouldn't this take the movie industry by storm. Unlimited detail (80Gb/s worth of detail)could be put forth into the models and cut rendering time. I'm not sure, any help here?

  100. Or maybe lss than.. by dilvish_the_damned · · Score: 1

    your average scsi3x raid0 array.

    --
    I think you underestimate just how much I just dont care.
  101. OK, but... by Boffin1 · · Score: 1

    ....do we really NEED to go this fast? "Won't someone PLEEASE think of the children?!"

  102. wow! by sewagemaster · · Score: 1

    >>transfer an unprecedented 80GB(!!) every second.

    my consumption of pr0n cant even go up to that high rate!!!

  103. Solid state means "no moving parts" by 0x20 · · Score: 1

    which sort of precludes MEMS...

  104. 224 GB is nothing by Florian+Weimer · · Score: 2

    It's only a fraction of the RAM size of an average supercomputer. No wonder Cray isn't at the top of supercomputing anymore, it's basically a reseller nowadays.

  105. sad by Anonymous Coward · · Score: 0

    that's pretty impressive. you want to know something equally as interesting, but far less of a "good thing" ...i' mgood friends with the guy that killed Semour Cray. when i found out it was my friend that killed him, i was going to kikc his ass... but he's a kempo teacher, so i didn't.

  106. Is this holostore technology? by serutan · · Score: 2

    About 10 years ago I read about a solid state device called a "holostore" that was in the prototype stage. It consisted of crystalline cylinders about a millimeter or so in diameter and 5 or 10 mm long, standing next to each other in a 50x50 array. Data was stored and read optically by three laser beams aimed at different angles polarizing the molecules, same principle as an LCD. I forget the total capacity but the transfer rate figure of 8 or 80Gb/second seems familiar. The prototype was said to have the same form factor as a 5-1/4-inch floppy drive, and like everything else it was supposed to be on the market in 2 or 3 years.

  107. Hang on a minute... by 2cool4school · · Score: 1

    In order for a transfer rate of 80 GB/sec to be useful doesn't RAM have to be capable of a higher transfer rate? Wouldn't you just get a huge bottleneck? Or is there some technology or use that this drive is put to that makes this irrelevant? If not, do Cray's computers have ram that is faster than 80GB/sec?

    Very impressive though.

    1. Re:Hang on a minute... by pkesel · · Score: 1

      You only need 80Gb/sec memory if there is only one memory path. If you have 20 paths and a controller to manage it you only need 4Gb/sec throughput, which isn't hard to come by.

      --
      - Sig this!
    2. Re:Hang on a minute... by SK-null · · Score: 1

      Last option

  108. eehhh... eehhh.... by Anonymous Coward · · Score: 0

    first post.

  109. Sorry dude, by ant_slayer · · Score: 1

    A Beowulf cluster implies commodity hardware (otherwise, it's a cluster-o'-workstations or a supercomputer). I don't think that this new technology comes anywhere close to your friendly neighborhood Wal-Mart shelf.

    -Josh O-

  110. The law "bytes == flops" by peter303 · · Score: 2

    Another variant of Moore's law (of who I forget stated it) says that a balanced supercomputer is "bytes == flops", that is it must be able to process as many bytes of data PER SECOND as it can do floating point operations. This has often meant that core memory must be about the same number of bytes as there are flops. This device goes a long way in satisfying this requirement.

    Early super computers in a generation sometimes skimp on such memory and are only good for problems that dont require much I/O like some physics simulations. Anything that processes data such as satellite imagery or seismic, weather, etc. requires significant memory capacity..

  111. No woman, no Cray by Anonymous Coward · · Score: 0

    *sob might as well get high*

    I'll never be able to afford some kewl tech like that..

    Anyways, we're jamming !

  112. HOLY POO! by Nobody's+Hero · · Score: 0

    This is impressive. With tranfers of 80 GB/S that is smokin'! when can i get me one of these for home? does anyone know the cost per byte on one of these bad boys??? Solid State is whoopass...all the flexibility and speed of DRAM with none of the data loss if you wind up with a poer outage..... Must HAVE!!!! *drool*

    --
    The Only Person Willing to be Me is ME!
  113. Great, But... by sirgoran · · Score: 1

    Can I get one for my home PC and will it cost less than the GNP of a small third-world country?
    (Just wondering)

    Goran

    --
    Carpe Scrotum - The only way to deal with your competition.
  114. RAM Highly useful for security by kannen · · Score: 1
    Well, actually, there's a very good reason to use RAM, and that's if you don't want to leave an electronic trail behind you after decrypting a file or whatever.

    If data security is important, then RAM is the way to go, because it is very easy to decisively destroy data that you don't want to be read. For that reason, there is a very big hubub about growing RAM storage sizes.

  115. Perfect... by superpulpsicle · · Score: 0

    Finally! Hardware good enough for my pR0n collection and distribute it at a fast rate.

  116. Um, by Anonymous Coward · · Score: 0

    how much does a pack of gum cost?

  117. Pr0n by Anonymous Coward · · Score: 0

    you know you're thinking it too.

    It's sad. but true.

  118. Finding a use for it by Theta116 · · Score: 1

    Bet the porn industry will find a use for the storage system

  119. Genome Compression by Suburban+nmate · · Score: 1

    Compressed genomes? Lossless I hope. I'd hate to see the compression "artifacts"...

    I'm no expert, but wouldn't it need a fat wedge of CPU cycles to [de]compress that data?

    That said, my 386SX 16Mhz (R.I.P) had a compressed disk, but even that was only about 2x the actual space, max.

    Theres no doubt about who they're marketing this to though. And these companies are filing patents on every genome they find, often without even knowing what they do! Not that it makes any difference whether they know or not, I DON'T WANT A PATENT ON MY /dev/body/*!! Thankyouverymuch.

    Ali

    --
    "Windows and Linux can co-exist on the same machine." - Microsoft Corporation.
  120. Oooo! Breaking CRAY SUPERCOMPUTER news... by Anonymous Coward · · Score: 0

    Uh, NEAT. Sure I'll trade-in my home PC for a CRAY SUPERCOMPUTER.....and gimmie one of them SPACE SHUTTLES too! Get real people...