Slashdot Mirror


The Ultimate All-In-One Storage Solution

karnifex writes "Filled up your LaCie Bigger Disk already, and looking for a little more storage space? Good news! The Petabox is ready! 'The petabox by the Internet Archive is a machine designed to safely store and process one petabyte of information (a petabyte is a million gigabytes).' And luckily, as the Internet Archive notes, it's shipping-container friendly (20' x 8' x 8'). So save on delivery costs and order two!"

387 comments

  1. Finally by foidulus · · Score: 4, Funny

    My million monkeys at a million terminals will have somewher to save all their potential Shakespeare works.
    But the question is, do my monkeys use VI or Emacs? That shall remain a mystery.

    1. Re:Finally by Anonymous Coward · · Score: 5, Funny

      Well, seeing as how monkeys type random gibberish,
      vi would seem like a perfect fit, yes?

    2. Re:Finally by Schaffner · · Score: 1

      In that case, TECO would be a better choice.

    3. Re:Finally by Anonymous Coward · · Score: 1, Interesting

      Real monkeys use ed. Anything else and its no monkey: it's a damned ape!

    4. Re:Finally by foidulus · · Score: 5, Funny

      A British researcher actually did trap some monkeys inside a room with a computer for a couple days. They flung large amounts of feces at the monitor and keyboard and beat the living crap out of the box, and typed the letter s a lot. Sounds like the response of a typical user upon their first vi experience.

    5. Re:Finally by Rodrin · · Score: 1

      Reminds me of my little cousins when they get to use the computer. Oh the memories. Now lets forget 'em.

    6. Re:Finally by MaxiCat_42 · · Score: 1

      Yes, you really need an editor that has a function to retrieve the settings of the front panel switches.

      Phil.

    7. Re:Finally by DrLZRDMN · · Score: 1

      I use IMPS

    8. Re:Finally by MaxiCat_42 · · Score: 4, Funny

      Sounds like the beginnings of Titus Andronicus.

    9. Re:Finally by Anonymous Coward · · Score: 1, Interesting

      Maybe after they focus on getting this done, they'll try to get the 1TB units fixed and shipped. My company was on backorder from the time they announced the 1TB unit (January) to the beginning of May. At this rate, the thing will available Q4. They have currently shipped 4 units thru PCMall/Macmall to date an the ones they have shipped are broken, hence the delay.

    10. Re:Finally by TTK+Ciar · · Score: 2, Funny

      Argh .. well, I made sure vi and emacs were installed, and jove is installed (my personally preferred editor), and pico is installed, but there is no teco.

      Guess your monkeys will have to use something else :-)

      -- TTK

    11. Re:Finally by Anonymous Coward · · Score: 0
      But the question is, do my monkeys use VI or Emacs?

      Two of the units to are required to store the Emacs binary.

    12. Re:Finally by cfl · · Score: 5, Interesting

      I just use this:

      Monkey Shakespeare Simulator

      Maybe not as much fun, but without the faeces
      I've noticed that Mozilla Firefox seems to give better results than IE

    13. Re:Finally by GunFodder · · Score: 1

      Trick question! Monkeys prefer the easy-to-use interface of Word.

    14. Re:Finally by CrazyTalk · · Score: 1

      Ob. Simpsons: "It was the best of times, it was the blurst of times???? Stupid Monkey!"

    15. Re:Finally by Anonymous Coward · · Score: 0

      Monkey's won't even use pico? Damn pico never gets the respect it deserves...

    16. Re:Finally by algae · · Score: 1

      Now the interesting thing is, one million monkeys at one million terminals (typewriters, whatever), will actually immediately start typing the complete works of Shakespeare.

      ...just not with all the letters in the right order (rim shot).

      --
      Causation can cause correlation
    17. Re:Finally by Ziviyr · · Score: 1

      Does Emancs have a monkey simulator to help fill that petathingy up?

      --

      Someone set us up the bomb, so shine we are!
    18. Re:Finally by pair-a-noyd · · Score: 1

      I can *personally* vouch for that....

    19. Re:Finally by gstoddart · · Score: 1

      It's also the response of us old timers who still use vi to most GUIs. :-P

      Trust me, there are no more icons that meaningfully convey what they signify/operate. I might eventually associate your damned icon with an action, but not in any meaningful way.

      And while we're at it, what's with all these weany vi clones that have modes that confuse even *me*? (and ":.,'a:s/^V^A/|/g" is a command I'm likely to use in a day)

      Young whipper snappers ... in my day, X-windows allowed you to have *three* command lines open at once, making you 12 times as productive. =)

      Hell, you could be both downloading *and* uudecoding pr0n while debugging code. *sheesh*

      --
      Lost at C:>. Found at C.
    20. Re:Finally by TheBoostedBrain · · Score: 1

      They'd use a Ximian Solution

      --
      -- When did Ignorance Become a Point of View?
    21. Re:Finally by masterQba · · Score: 1

      with a couple of these babies my life long dream to mirror the internet is possible.

      --
      xb0x
    22. Re:Finally by Wolfrider · · Score: 1

      --Hmm... Petabyte. I wonder how many Tamagotchis and Commies that will hold...

      ~;-)

      --
      .
      == WolfriderV6 == I'm willing to admit that *I just might* be wrong... Are you??
    23. Re:Finally by Anonymous Coward · · Score: 0
      Hell, you could be both downloading *and* uudecoding pr0n while debugging code.

      Nah, use the third window to view the porn with 'less', then you can scroll in both directions to see the ASCII images.

    24. Re:Finally by CvD · · Score: 1

      You owe me a new keyboard. :-) And some way to clean up my monitor. Hilarious!

    25. Re:Finally by Anonymous Coward · · Score: 0

      Interestingly (or not) the UK band Chumbawumba got their name from a monkeys typing experiment. It was the only readable word that they typed.

      Or so they claim...

    26. Re:Finally by Anonymous Coward · · Score: 0


      well they managed to get 15 letters from "Pericles" after 958,399,000,000,000,000,000 monkey-years !.

      Sent in by Ken Phillips from Roseville, California on 12 April 2004.
      "[Enter GOWER.] ?2IDzPN9sq6V ;e'?nGI3&?3 La""0 ..."
      matched
      "[Enter GOWER.] [Before the palace of Antioch.] To sing a song that old was sung, From ashes ancient Gower is come; Assuming man's infirmities, To glad your ear, and please your eyes."

    27. Re:Finally by Paulrothrock · · Score: 1

      What's wrong with that. I fling poo at the torture device in my cubicle daily.

      --
      I'm in the hole of the broadband donut.
  2. colossal... by ambienceman · · Score: 2, Funny

    Imagine all the Spice Girls' songs you could save on that thing...wow...

    1. Re:colossal... by foidulus · · Score: 4, Funny

      Don't just stop at songs! Store the DNA sequences for the spice girls on it, such that years from now our decendants may know the joys of Scary Spice.

    2. Re:colossal... by ambienceman · · Score: 2, Interesting

      Yeah and maybe we could clone a few of them (a couple scary and sporty spices) for our own conveniences.....hehe.for singing of course

    3. Re:colossal... by RumpRoast · · Score: 0

      ok, now THATS funny.

      --

      My Ass hurts.
    4. Re:colossal... by Cyno01 · · Score: 3, Funny

      They sing?

      --
      "Sic Semper Tyrannosaurus Rex."
    5. Re:colossal... by glwtta · · Score: 4, Informative
      Store the DNA sequences for the spice girls on it, such that years from now our decendants may know the joys of Scary Spice.

      The current genome build has a size of 3,020,300,000 bp, at 2 bits per bp and 5(?) spice girls, that's about 3.5 GB (uncompressed).

      Of course with a mostly static database like that you only want to store the diffs, not the whole thing. The bulk of the diff would be SNPs, roughly 1 per 1000 bp: 3,020,300,000 / 1000 / 4 / 1048576 that's about 0.72MB per spice girl. An if you only store the ones actually different from wildtype you probably don't need more than 20% of that.

      You can fit a Spice Girl on a floppy.

      --
      sic transit gloria mundi
    6. Re:colossal... by iminplaya · · Score: 4, Funny

      You can fit a Spice Girl on a floppy.

      Or, you can fit two floppies on a Spice Girl
      "Sock it to me."

      --
      What?
    7. Re:colossal... by theTerribleRobbo · · Score: 1


      You'd have to wait 20 years for them to grow up first, though.

    8. Re:colossal... by ngoy · · Score: 1
      Or, you can fit two floppies on a Spice Girl
      I think most of the /.'s on here are thinking of fitting their hard drive in a Spice Girl.
      --
      --ngoy
    9. Re:colossal... by guacamolefoo · · Score: 1

      I think most of the /.'s on here are thinking of fitting their hard drive in a Spice Girl.

      Nothing like modding a box for fun...

      GF.

    10. Re:colossal... by Anonymous Coward · · Score: 0

      Wouldn't that make them sp l ice girls?

    11. Re:colossal... by sepluv · · Score: 1
      thinking of fitting their hard drive in a Spice Girl
      Surely you mean their massive 1-PB 20-foot hard dic^Hsks?
      --
      Joe Llywelyn Griffith Blakesley
      [This post is in the public domain (copyright-free) unless otherwise stated]
    12. Re:colossal... by Anonymous Coward · · Score: 0


      While you can fit a Spice Girl on a floppy, she won't enjoy it much. Luckily I just recieved several emails saying they can take care of that annoying "floppy" problem with Herbal Viagra.

  3. I'd buy one.... by david_reese · · Score: 5, Funny
    but I heard that all you can store on there are
    ...drumroll

    Peta-files

    1. Re:I'd buy one.... by Smidge204 · · Score: 5, Funny

      I heard the FBI managed to squeeze 1000 of these into their annual budget...

      Exa-Files.
      =simdge=

    2. Re:I'd buy one.... by BinLadenMyHero · · Score: 1

      Not peta-files, but one peta-file only.
      The device has storage for only one peta-bytes.

    3. Re:I'd buy one.... by shigelojoe · · Score: 1

      *Waves hand over head*

      "Whoosh!"

    4. Re:I'd buy one.... by mrwonton · · Score: 1

      Are you sure you can't store 8 of them then?

      --
      Not more than you need, just more than you want
    5. Re:I'd buy one.... by Atticu5 · · Score: 1

      No, those are peta-bits.

  4. Can you get one... by hemp · · Score: 2, Funny

    if you still enjoy wearing furs??

    --
    Skip ------ See the latest from http://www.anArchyFortWorth.com
    1. Re:Can you get one... by KevinKnSC · · Score: 1

      Yes, but in the middle of the night it will let all of your data out and scratch up the side of your SUV.

  5. Applications by kevmo · · Score: 0

    I can think of a few good applications for this much storage ...

    1. Re:Applications by Anonymous Coward · · Score: 0

      Well, FOR GOD'S SAKE, kevmo, don't keep us in the dark!!!!!! What can you use it for?!?!?!!? ;)

  6. Business idea by j_hirny · · Score: 2, Interesting

    Maybe someone should try to sell these boxes to GMail? They will surely need a lot of storage space.

    1. Re:Business idea by Anonymous Coward · · Score: 0

      I can assure you google already has more stroage avilable than one of these boxes.

    2. Re:Business idea by timeOday · · Score: 2, Insightful
      I suspect google is more interested in building a platform that would be a competitor to this product. For the device in this article they estimate 1 FTE (full-time-employee) for each petabyte of storage. That doesn't sound so good. Google's system will apparently replicate and migrate data betwen units as necessary so you never need to replace drives at all; the maximum capacity just degrades slowly with time. Perhaps when it gets to 80% original capacity you just roll in a newer unit (which is probably much bigger in capacity as well), hook them together for a day, then throw out the old one.

      The power requirements are also quite hefty. It shouldn't be necessary to run all those drives (and the computers behind them) unless the unit is near capacity and access is random (which I'm sure would rarely be the case). Instead, they should be dynamically powering drives and computers up and down, and migrating data to a reasonably small 'working set' of drives.

      On the hardware front, the device in this article also incorporates 800 "low-end PCs." IOW it's a big cluster that happens to be heavy on storage. If all you want is the storage, surely there is some way to get rid of all those motherboards and CPUs with their fault-prone, power-hungry fans. They need to develop a controller that can directly handle, say, 64 hard drives, analogous to a big network switch.

      Anyways, it sounds like a fun project!

    3. Re:Business idea by AlecC · · Score: 2, Informative

      They need to develop a controller that can directly handle, say, 64 hard drives, analogous to a big network switch.

      It's called a Fibre Channel controller. Fibre Channel loop (which disks use) offers a total of 255 addresses - which has to include the controller. Disks now available in the 300Gbyte region, so 80 Tbyte/loop seems reasonable (and, according to the article, they seem only to have 100Tbyte up so far). 12 of these loops will give you your petabyte. Mind you, you will waste the disk bandwidth; this will gicve you capacity but not throughput.

      --
      Consciousness is an illusion caused by an excess of self consciousness.
  7. Price? by llin · · Score: 1

    I didn't notice a price specified. Am I just not seeing it? Anyone in a position to give details?

    1. Re:Price? by Anonymous Coward · · Score: 3, Insightful

      If you have to ask, you can't afford it. Just remember that. It might come in handy again someday. :)

    2. Re:Price? by Anonymous Coward · · Score: 0

      This mentions $130k/rack.

    3. Re:Price? by RollingThunder · · Score: 4, Informative

      In the "discussion" blocks down below there's a price link.

      Rack materials cost is currently estimated to be $121K for 96TB. Node materials are a just under $1450. This price does not include markup, assembly or burn-in from the system integrator and thus will increase by another 5-7% to approximately $130K/rack.

      The weight of a fully-loaded rack is estimated to be 1500 lbs. That figure may rise depending on what hardware is required for rack cooling.

      Power is estimated to be 5500 watts. This too will depend on rack level cooling equipment.

      These figures assume no external 1G Ethernet NICs.

      For a breakdown of all the above, see the attached spreadsheet.

    4. Re:Price? by cliveholloway · · Score: 2, Informative

      From the forum:

      Rack materials cost is currently estimated to be $121K for 96TB. Node materials are a just under $1450. This price does not include markup, assembly or burn-in from the system integrator and thus will increase by another 5-7% to approximately $130K/rack.

      So, about $1.3M (10 racks)

      cLive ;-)

      --
      -- Trinity in high heels carrying a whip: The donimatrix - there is no spoonerism
    5. Re:Price? by theLOUDroom · · Score: 2, Insightful

      So, about $1.3M (10 racks)

      What would be interesting is to know the estimated maintenance costs as well. With than many drives, I imagine you'd be changing them like light bulbs, especially as time passes and the probability of each drive failing get's higher and higher.

      If one was really clever, they could use the failure rate of a typical hard disk and Moore's Law to estimate monthly replacement costs for the next 100 years or so. I would expect them to rise in the short term as the drives age, but fall in the long term as moore's law catches up.

      --
      Life is too short to proofread.
    6. Re:Price? by I(rispee_I(reme · · Score: 1

      I'm pretty sure that Moore's law only applies to cpu speed, not hard drives.

    7. Re:Price? by theLOUDroom · · Score: 1

      I'm pretty sure that Moore's law only applies to cpu speed, not hard drives.

      And I'm pretty sure that Moore's Law is not actually a "law". Anyways, it seems to fix pretty well for hard drives, as it has for transistor counts/clock speeds.

      It doesn't makes sense to get picky about Moore's law when it was nothing but a SWAG anyways.

      --
      Life is too short to proofread.
    8. Re:Price? by trentblase · · Score: 1
      If you have to ask, you can't afford it. Just remember that. It might come in handy again someday. :)

      Insightful? BS! I've had to ask for the price of tons of stuff that I could afford. Or are you using some highly specialized definitions of "had to" and "afford"? Just because I can afford something doesn't mean I think it's a good enough value to purchase the item.

    9. Re:Price? by HTH+NE1 · · Score: 1

      So, about $1.3M (10 racks)

      And if you can afford that, I really don't think you'd want to back it up to CD-Rs.

      --
      Oh, say does that Star-Spangled Banner entwine / The myrtle of Venus with Bacchus's vine?
  8. One question... by Cyno01 · · Score: 2, Funny

    How many LOCs/VW Bug?

    --
    "Sic Semper Tyrannosaurus Rex."
  9. dongle....? by Anonymous Coward · · Score: 0

    I'm skeptical of any promotional website that uses the word "dongle".

    haha... dongle

  10. Petabox is ready! by KevinKnSC · · Score: 5, Informative
    Good news! The Petabox is ready!

    From the article:
    PILOT STATUS 5/2004
    * The first 100TB Rack is up and running!
    * The second 100TB Rack will be up by the end of May

    Apparently this is some new use of the word "ready" with which I am not familiar. Neat technology, no doubt, but it doesn't really look like it's ready for prime time just yet.

    1. Re:Petabox is ready! by RollingThunder · · Score: 1

      It's not all that usual, true - most "ready" devices these days don't provide even a partial working unit, just finalized specs. ;)

    2. Re:Petabox is ready! by DAldredge · · Score: 1

      Please explain.

    3. Re:Petabox is ready! by Anonymous Coward · · Score: 0

      That's why I'm voting for Reagan!

    4. Re:Petabox is ready! by joebeone · · Score: 1

      I saw this recently and their news site is not up to date... they had one 2-rack, 250 TB box up and running... three more of these suckers and you've got a PB. The hard part, by far, was making sure that they could put two of these racks together without starting a fire...

    5. Re:Petabox is ready! by meadowsp · · Score: 1

      I wouldn't speak too soon, Bush may well end up eating some cheese before too much longer.

  11. In 10 years ... by Bob+Loblaw · · Score: 5, Insightful

    Will we find one of these things in eBay in 10 years selling for $10 and feel all nostalgic about those days when that amount of storage media was the size of a room?

    1. Re:In 10 years ... by DrEldarion · · Score: 4, Insightful

      It does make you nostalgic thinking about things like that. I have more storage on a tiny card in my digital camera right now than every computer in the computer lab at my Jr. High had combined. My cell phone is more powerful than my first desktop computer. I can download in 5 minutes what would have taken me a month to download back then. Ah, technology.

    2. Re:In 10 years ... by Anonymous Coward · · Score: 0

      Ten years ago, I think big hard drives were around one gig. Now, they're around 300 gig. If they keep growing at the same pace, we'll see one petabyte hard drives in 14.2 years.

    3. Re:In 10 years ... by Anonymous Coward · · Score: 0

      There was a time a few years back when my PDA was considerably more powerful than my Dad's desktop. He'd fallen behind a bit...

    4. Re:In 10 years ... by slickwillie · · Score: 1

      In 10 years electricity will be so expensive that it will not be economically feasible to run your laptop, much less this monster.

      BTW, does anyone else use Google as a spell checker? I just brought up a Google page, typed in "feasable" and it replied back "Did you mean feasible?".

    5. Re:In 10 years ... by Anonymous Coward · · Score: 0

      Oh yeah. I used to use word to spell check single words, but google is my new best friend.

    6. Re:In 10 years ... by blancolioni · · Score: 2, Insightful

      Every hard drive I've ever bought has been larger than all my previous hard drives combined. And this is without even trying.

      The storage problems I have these days are almost entirely organisational.

    7. Re:In 10 years ... by jafuser · · Score: 1

      I have more storage on a tiny card in my digital camera right now than every computer in the computer lab at my Jr. High had combined.

      I saw one of these at a computer store a few days ago and was kind of surprised how much storage could be put in such a small package. The card is 20.0 x 25.0 x 1.7mm in size and holds 512MB, with a 1GB version coming soon.

      --
      Please consider making an automatic monthly recurring donation to the EFF
  12. Hmmm by Anonymous Coward · · Score: 0

    How many petadollars does it cost?

  13. Don't get too excited by 42forty-two42 · · Score: 3, Informative

    They only have one rack, which is 100 TB.

    1. Re:Don't get too excited by danormsby · · Score: 4, Funny

      Almost a Petabit though.

      --
      Omnis amans amens
    2. Re:Don't get too excited by 0racle · · Score: 2, Funny

      Ya but what a rack it is.

      --
      "I use a Mac because I'm just better than you are."
  14. To give you an idea of how much that truly is: by Three+Headed+Man · · Score: 5, Interesting

    If you gave me a 100 mbit line, it would take me over 92 days to fill it up with porn. More if I slept.

    --
    I'm probably at the karma cap. Mod up a funny troll instead, it lightens the mood :)
    1. Re:To give you an idea of how much that truly is: by Bob+Loblaw · · Score: 1

      You are obviously a pr0n novice ... true professionals would do it in a fraction of that time.

    2. Re:To give you an idea of how much that truly is: by Anonymous Coward · · Score: 0

      You're both obviously novices. I already have a petabyte of pr0n!

    3. Re:To give you an idea of how much that truly is: by Three+Headed+Man · · Score: 1

      You're right, a pro would have faster internet, but the point I was trying to make was that on what most people would consider fast internet right now, the physical limit would make it take over 3 months to fill up the storage with data.

      --
      I'm probably at the karma cap. Mod up a funny troll instead, it lightens the mood :)
    4. Re:To give you an idea of how much that truly is: by ecampbel · · Score: 4, Informative

      You're off by a factor of ten. Let google do the math: 1 petabyte / 100 megabits / second in days = 994.205393 days

      --

      Sig goes here
    5. Re:To give you an idea of how much that truly is: by christopher240240 · · Score: 1

      I invented a program that downloads porn from the internet one million times faster.

    6. Re:To give you an idea of how much that truly is: by Anonymous Coward · · Score: 0

      That would be 100 tits per second.

    7. Re:To give you an idea of how much that truly is: by zyridium · · Score: 1

      And they have only built 1/10th of it so far...

    8. Re:To give you an idea of how much that truly is: by perky · · Score: 1

      oooooh. more google whizziness. apprepreciate the pointer.

      --
      "The new wave is not value-added; it's garbage-subtracted" - Esther Dyson, Dec 1994
    9. Re:To give you an idea of how much that truly is: by Anonymous Coward · · Score: 0
      More if I slept.

      Sleep? Pshaw. Use Lesbian GNU/Linux and a cron job and you can have the computer work around the clock for you.

      #!/bin/sh
      cd ~/.pr0n/
      ./porn-get update
      for a in `./porn-cache search |awk '{print $1}'`; do
      ./porn-get install $a
      done

      Just have that run as often as necessary and you can get all the sleep you want... :-)

  15. holy cow by Anonymous Coward · · Score: 0

    something that can finally hold all of my porn

  16. Potential customers by thedillybar · · Score: 4, Funny
    "After extensive market analysis, we have found 1 (one) organization that is interested in purchasing this device."

    Can we say, Goooooooooooooooogle?

    1. Re:Potential customers by Short+Circuit · · Score: 1

      I didn't see throughput stats on the box. I don't know if it'd be all that useful for Google, if the box can't keep up with the load.

    2. Re:Potential customers by Anonymous Coward · · Score: 0

      "After extensive market analysis, we have found 1 (one) organization that is interested in purchasing this device."

      Can we say, Goooooooooooooooogle?


      Well actually, CERN could use a few dozen, when the LHC comes online. They need petabyte data stores as cache.

    3. Re:Potential customers by curator_thew · · Score: 1

      "After extensive market analysis, we have found 1 (one) organization that is interested in purchasing this device."

      There's probably also only a market for about 4 computers in the world.

    4. Re:Potential customers by Anonymous Coward · · Score: 0

      I recall an article ~10 years ago regarding Exxon's storage solution for offshore seismic survey data. They stored multi PB's of data on digital video tape with automatic tape changers. Probably next to the Lost Arc in a big warehouse. You know you are bad when your storage abilities are still mind boggling a decade later...

    5. Re:Potential customers by moosesocks · · Score: 4, Interesting

      I believe that back in the 50s, the president of IBM enthusastically proclaimed that there was potentially a worldwide market for four, possibly even five computers. And this was good news.

      So don't laugh!

      (I'm sure there are PLENTY of organizations which could use this type of storage. The IRS and NASA being among them)

      --
      -- If you try to fail and succeed, which have you done? - Uli's moose
    6. Re:Potential customers by lewp · · Score: 4, Funny

      NASA just uses an alien with a really good memory to store all their data. The IRS doesn't store any data at all. They just make shit up on the fly.

      --
      Game... blouses.
    7. Re:Potential customers by bucky0 · · Score: 1

      The IRS doesn't store any data at all. They just make shit up on the fly.

      I knew it!!!

      --

      -Bucky
    8. Re:Potential customers by MikeXpop · · Score: 1

      I remember that was 1948. And I think that quote appeared in popular mechanics, though I'm not quite sure.

      I also remembering something about it mentioning computers may weigh less than a ton in the future, though I may be mixing references here.

      --
      Etiquette is etiquette. He kills his mother but he can't wear grey trousers.
    9. Re:Potential customers by moosesocks · · Score: 1

      Ah. Finally found it. Thanks for the Pop. Mechanics refrence!

      It was in 1943.

      "I think there is a world market for maybe five computers." -- Thomas Watson of IBM

      --
      -- If you try to fail and succeed, which have you done? - Uli's moose
    10. Re:Potential customers by Feyr · · Score: 1

      i think that was DEC's president, not IBM's

    11. Re:Potential customers by pair-a-noyd · · Score: 1

      And you can bet your ass Admiral Poindexter and TIA are one the list to get a few of these...

    12. Re:Potential customers by Tore+S+B · · Score: 1

      I believe that back in the 50s, the president of IBM enthusastically proclaimed that there was potentially a worldwide market for four, possibly even five computers. And this was good news.

      You have to think about what computers were being used for back then. This was not stupidity on his part, but an accurate analysis of how many computers would be needed for the work pattern that computers were being used with back then. The machines would probably be batch-processed, and everyone would log in via remote terminals and put their batch of cards on the remote reader.

      --
      toresbe
    13. Re:Potential customers by AlecC · · Score: 1

      No, DEC's supremo Ken Olsen said that he could see no need for a computer in the home. The IBM quote is correct.

      --
      Consciousness is an illusion caused by an excess of self consciousness.
    14. Re:Potential customers by Anonymous Coward · · Score: 0

      The IRS doesn't store any data at all. They just make shit up on the fly.

      So they use monkeys ?

  17. Wayback machine by DigiShaman · · Score: 1

    I find it ironic that the Wayback machine poster is right next to the unit. Do I detect a hidden message? Hmmmm ;)

    --
    Life is not for the lazy.
  18. LaCie disk by Anonymous Coward · · Score: 0

    ok, what do you do if you want to use that Lacie disk and share it between a system running WInXP and another running OS X or linux?

    any ideas?

    1. Re:LaCie disk by isorox · · Score: 1

      fat32, one partition :D

    2. Re:LaCie disk by Anonymous Coward · · Score: 0

      OS X can access fat32?

    3. Re:LaCie disk by arkhan_jg · · Score: 1
      fat32, one partition :D

      Imagine trying to defrag it though...

      --
      Remember kids, it's all fun and games until someone commits wholesale galactic genocide.
    4. Re:LaCie disk by isorox · · Score: 1

      Yes

    5. Re:LaCie disk by justMichael · · Score: 1
      OS X can access fat32?

      Apparently it's been possible since OS 9, it's about half way down.
    6. Re:LaCie disk by Lando+Griffin · · Score: 0

      That chart says OS9 can't connect to FTP servers?!?!

  19. Dupe by Patik · · Score: 0, Offtopic
    1. Re:Dupe by isorox · · Score: 1

      This story is about 1PB of storage, the Lacie is old news, not a dupe

  20. Re:wrong by cybermace5 · · Score: 1

    What's a terabyte, then?

    --
    ...
  21. I am NEVER going to get caught up... by ackthpt · · Score: 2
    I swear, I can't keep up. I just got myself up to 1/2 terrabyte and could easily get to a full terrabyte, then this petabyte sh!t comes along.

    Ah well, all I'd do is fill it up with Simpsons episodes....

    --

    A feeling of having made the same mistake before: Deja Foobar
    1. Re:I am NEVER going to get caught up... by ldm314 · · Score: 1

      Every simsons episode only ammounts to about 50 GB as mpeg VCD quality. I assume you are wanting HD quality and then you'd probably be able to fill it.

  22. Ummmm by akira69 · · Score: 2, Funny

    20 feet by 8 feet by 8 feet eh.... I thought they had that in the 1970s

    1. Re:Ummmm by Anonymous Coward · · Score: 3, Funny

      "20 feet by 8 feet by 8 feet eh"

      In 2010 it better be 1 by 4 by 9.

    2. Re:Ummmm by cujo_1111 · · Score: 1

      heheheh

      Nice work... mod parent to +5, Obscure movie reference

      --
      If I point out that you are incorrect, making me a foe does not make you any more correct.
    3. Re:Ummmm by shigelojoe · · Score: 2, Funny

      Obscure? This is Slashdot we're talking about here. ;D

  23. Re:wrong by j_hirny · · Score: 1

    Since when? 1Tera is 1,000 Giga and 1Peta is 1,000 Tera.

  24. Re:wrong by ditto999999999999999 · · Score: 1

    No, a terabyte is a thousand gigs, a petabyte is a million.

  25. Finally... by homeobocks · · Score: 0

    ...storage large enough that in 4 years, everyone will have the terabyte of secondary memory needed for Longhorn!

    --
    MOUNT TAPE U1439 ON B3, NO RING
    1. Re:Finally... by RyatNrrd · · Score: 1

      So jokes can't be funny if they're about Microsoft? Ooh, shit, sorry - is it too soon?

  26. Re:wrong by Vihai · · Score: 3, Funny


    Wrong :)

    One PetaByte is 1,000 TeraBytes which are 1,000,000 GigaBytes wich are 1,000,000,000,000,000 Bytes wich are 3.35 LotsOfPr0n.

  27. Re:wrong by KevinKnSC · · Score: 1

    Sorry, a terabyte is 1000 gigabytes, and a petabyte is 1000000 gigabytes. The article was correct.

  28. Sooooo.... by Boogaroo · · Score: 2, Interesting

    Think someone like the government is going to keep track of who buys these things?

    1. Re:Sooooo.... by Anonymous Coward · · Score: 0

      That'll be easy, just use a gravitational anomoly detector.

    2. Re:Sooooo.... by TTK+Ciar · · Score: 1

      > Think someone like the government is going to keep track of who buys these things? They can't. The complete specs for making your own will be published (RSN), as well as the software and documentation, all open-sourced. Anyone will be able to build one without going through us at the Archive. -- TTK

    3. Re:Sooooo.... by Anonymous Coward · · Score: 0

      Think someone like the government is going to keep track of who buys these things?

      Absolutely. That's why they install radio transmitters in your teeth when you get your fillings: They activate the device when they find out you've bought too many hard drives. Then they start drugging your food to prevent you from accumulating 100 TBs of data and uncovering their nefarious plans to suppress Ogg Vorbis.

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

      I think anyone with any rack mounting experience could already do this without the plans.

      Here's the preliminary plans:

      1) Find the cheapest 1U server that has a maximum amount of disk space.
      2) Buy as many as you can afford or need for your storage requirement (which ever comes first).
      3) Install in racks and turn on.
      4) Upgrade power in room and reset breakers.
      5) Upgrade cooling in room and reset fire suppression system.
      6) Start to collect data.
      7) How do we back it all up?
      8) ???
      9) Profit

  29. Re:wrong by Smidge204 · · Score: 1

    No... 1,000 Gigabytes is 1 Terabyte. 1,000 Terabytes is one Petabyte. Therefore, 1,000,000 Gigabytes = 1 Petabyte.

    Specifically, 1,048,576 GB
    =Smidge=

  30. wrong by ResidntGeek · · Score: 1

    1 petabyte is 1,000 terabytes, not gigabytes...

    --
    ResidntGeek
  31. Ahhh... by 10101001011 · · Score: 2, Funny

    I just updated my old stone disk (it was erroding) with the latest gerbil in a wheel (I couldn't afford the guinea pig) with awesome seek times (he can seek food in less than 30 nanoseconds) and at least double-digit RPMs.

  32. Mark this moment by yebb · · Score: 1

    a few years from now when we have pentabyte hard drives selling for less than the cost of visit to the local grocery store. Look back at this article and say, "I remember when..."

    1. Re:Mark this moment by jrockway · · Score: 1

      Pentabyte hard drives? I think I have quite a few of those around (pentabyte = 5 bytes)

      --
      My other car is first.
  33. Backups by Anonymous Coward · · Score: 0

    What do I backup my petabox with?

    1. Re:Backups by TTK+Ciar · · Score: 1

      > What do I backup my petabox with? Under the current setup, the petabox backs itself up onto "partner" nodes, so if a disk drive bursts into flame, a new one can be plugged in and refilled with its old data. -- TTK

  34. Ohh Good, now I can buy Longhorn by l0ungeb0y · · Score: 4, Funny

    Though it says Linux is the standard OS, I'm hoping they plan on optimizing for Longhorn... so far this is the only system out that can meet Longhorns recommended disk capacity and RAM requirements. ...now if they could only find a way to fit all that into a mini-ATX tower.

    1. Re:Ohh Good, now I can buy Longhorn by Anonymous Coward · · Score: 0

      That was so funny, yet I have no Overrated mod points to hand out...

  35. LaCie by Short+Circuit · · Score: 1

    If you click on the link for the LaCie disk, you'll see, at the bottom of the page, that it won a 4.5 rating at Macworld in June, 2004.

    Did I miss a month in there somewhere?

    1. Re:LaCie by Anonymous Coward · · Score: 0

      That issue is done, but perhaps it isn't for sale yet?

  36. Uber-huge! by Enonu · · Score: 1

    20' x 8' x 8' = 1280 cubic feet!

    Jesus, that's more volume than the average room in a house! What year is it, 1984?

    1. Re:Uber-huge! by DrLZRDMN · · Score: 2

      I think your confusing square footage with cupic footage, if the average house had ten foot high ceilings, it would only take up 16 feet of floor space

    2. Re:Uber-huge! by vlag · · Score: 2

      I think you're confused. A normal 2 bedroom house (that doesn't have wheels) will have a floor space around 1600-1700 square feet. With 8 foot ceilings, the interior (3D) space would be about 12,000 cubic feet at least.

      --
      Do you want to remove linux?
    3. Re:Uber-huge! by zeno_2 · · Score: 1

      Or if you want to be accurate its going to take up 160 sq. feet of floor space.

    4. Re:Uber-huge! by Anonymous Coward · · Score: 0

      If you want to be really accurate... it would be 128 square feet... (20 x 8 x 8 = 1280 and 1280 / 10 = 128)

    5. Re:Uber-huge! by platipusrc · · Score: 1

      so...if you had 10 rooms, the average room would be what, about 1200 cubic feet?

      --
      And the muscular cyborg German dudes dance with sexy French Canadians
    6. Re:Uber-huge! by zeno_2 · · Score: 1
      If you want to be really accurate... it would be 128 square feet... (20 x 8 x 8 = 1280 and 1280 / 10 = 128)

      Hmm thats interesting math.

      As far as I understood, an object who's depth is 20ft and width is 8ft, would take up 160sq. feet of floor space. Im not even sure why your taking into account the height of the object, or why your dividing by 10...

      That might just be me though..

  37. two words by Loconut1389 · · Score: 4, Interesting

    Good God.

    or alternatively

    What for?

    At least as far as the next year or two is concerned. RIAA has all but outlawed music on the computer and even so, a petabyte of $1.25 songs would cost you more than bill gates makes in a year. If you have a petabyte of home movies, you must be making porno films.. If you have a petabyte of DVD's ripped, you have several life sentences coming, even if you own all the dvd's somehow (more bill gates salary multiples). And if you have text files, then holy grapes batman, youll never read all that in 10 lifetimes.

    I can see uses in the comercial realm, buying multiple units in order to backup. But if this is in anyway marketed toward the consumer, only the biggest 'mine has to be bigger than yours' geek would buy something like that right now. I'll probably have one of those on my desk/floor about 5 to 7 years from now when its affordable/realisitic for me.

    1. Re:two words by periol · · Score: 1

      please. it's called netflix and uncompressed dvd rips.

    2. Re:two words by Dog135 · · Score: 1

      Home movies of course!

      Imagine: Your wife decides one day that she's going bi. She'll be having many women over for extended "sleep overs" and will let you video tape them. You go out and buy a few dozen HDTV digital video cams and need soemthing to store all that video on.

      Simple. Happens all the time.

      --
      "That's so plausible, I can't believe it!" - Leela
    3. Re:two words by pbox · · Score: 5, Insightful

      Assuming 2 layered disks that is 10 GB per disk (feeling generous).
      100 disk -> 1 TB
      15000 disks -> 150 TB.

      Netflix has a "mere" collection of 15000 disks. Your patebyte disk is only 1/6th full.

      You upload all music CDs: 1 GB per disk (feeling generous).

      How many CDs can be in print? Maybe a 500,000?

      That is only 500 TB. Now your disk is 2/3rd full.

      Lets upload all printed material. May or may not fit in the rest.

      Then again, if you want to archive the internet: ~6G pages. 10kB each. 60 TB. each run. Store the last 16 versions -> 1TB.

      --
      Code poet, espresso fiend, starter upper.
    4. Re:two words by TTK+Ciar · · Score: 1

      Well, at the Archive we host "contribution" file sets (basically enduser-submitted audio, text, and video documents which are not protected works), and we also crawl the known world wide web about once every two months or so (about 60TB worth of data) and save it. This first petabox is being pre-stocked with our complete "collections" data (about 25TB worth) and a part of the "web" data. It's not *quite* the authorative repository of all human knowledge :-) but it's a start. You could think of it as a huuuuuuge web browser cache ;-) -- TTK

    5. Re:two words by Anonymous Coward · · Score: 0

      you'd think they'd compress the pages or use a versioning system like CVS or SVN instead of 16...

    6. Re:two words by TTK+Ciar · · Score: 1

      To give you some idea, the text (and only text, not pictures, and in plain vanilla ascii format) of the Library of Congress will fit in 1/4 of one rack (20 storage nodes). So figure a ten-rack PetaBox will store 40 Libraries of Congress. I don't know about you, but I'll take good, dense reading material over movies or music any day. Of course my preferences are low on the priority list :-) so it's being pre-filled with all kinds of flashy stuff. -- TTK

    7. Re:two words by Loconut1389 · · Score: 1

      I understand the Internet Archives use (Im a huge fan of the wayback machine). I guess I'm mostly trying to say that if a home user had/could afford one of these like the way the original poster implied, you'd have to be crazy or in jail =)

    8. Re:two words by Malawar · · Score: 0

      If I remember correctly (my memory sucks), the Internet Archive is building this, because, they are.. wait for it... Archiving the internet! That could take a hefty bit of space.

    9. Re:two words by pbox · · Score: 1

      I like reading, Library of Congress gives a pretty good selection, IMHO.

      Anyway then we can conclude that almost all movies on DVD, all CDs and all book in the LoC fit on this PetaBox. Probably would fit a good chunk of all published photographs as well. That is a good start for a Media Center.

      But (as you can probably tell I am reading William Gibson), it would only fit a small portion of a complete sensory input. That bandwidth must be ginormus. To record that even in compressed mode must take multiple petabytes.

      Even what we can record:

      visual: about 5GB/hr in crappy over-compressed DVD native format.
      audio: included above
      temperature (touch): probably less than 1K/hr
      pressure (touch): ??? probably less than 100MB/hr
      I am too lazy, but body-surface/pressure-sensor-area * pressure-range/pressure-sensitivity * pressure-sensivity-dynamics-time * 1 hour
      smell/taste: can't really measure...

      So we process 5.1GB+ per hour, probably even more (much more).

      So in a lifetime of 80 years we percive (as we can record):

      3.6 PB (peta).

      Conclusion: you just need 4 of these PetaBoxen and you can make a complete DVD of your life 24/7.

      Given that Moore's Law seems to underestimate the storage size growth (they doubled every 12 months, instead of 18).

      2004: 250GB ... (left to reader's excercise)
      2018: 4PB

      So by that time we can have a single harddisk containing the complete "DVD of Your LifE (TM)" or DOYLE for short.

      BUT WHO THE HELL WOULD WANT TO WATCH THAT BORING SH!T???

      --
      Code poet, espresso fiend, starter upper.
    10. Re:two words by Anonymous Coward · · Score: 0

      Moore's law has serious problems. It's more of a theory than a law, because there is no proof, just correlative data. THis is how politicians lie. Let's not make ourselves as low as them

      (witty remark here)

    11. Re:two words by maczmail · · Score: 2, Funny

      Good luck trying to create a taxonomy to organize a PB worth of objects much less a file system! /dev/FrigginHUGE

    12. Re:two words by Twirlip+of+the+Mists · · Score: 1

      RIAA has all but outlawed music on the computer

      I'm startled to hear that, considering that my music library includes more than 11,000 tracks and 44 GB. Closer to 90, because I keep a redundant copy on a separate hard drive.

      Pirated music is not okay, of course, but the RIAA didn't have anything to do with that. It was illegal a long time before the RIAA came along.

      Look, if you want to fill up a petabyte, just do this: never throw anything away. Ever. Every CD you've ever bought, every DVD, every piece of software you've ever purchased. Every phone number, every vacation photo, every email, every phone call. Every bad poem you ever wrote. Every love letter you ever received. Every home movie of your kids you ever shot. Keep it all on line, forever.

      Would that fill up a petabyte? No, probably not. But for a family of four....

      A hundred years ago, before we had home movies and vacation photos, it was the family bible. In it was recorded every milestone in every person's life: births, deaths, marriages, whatever. Family bibles were passed down for generations.

      A couple decades from now, it's gonna be the family hard drive.

      Don't ever throw anything away. Keep it, pass it down. Be the archivist of your own personal history.

      (Adding the layers of intelligence that can keep an archive like that in a semblance of order is a challenge for the programmers out there. Go be clever.)

      --

      I write in my journal
  38. Not really a Petabyte...yet by Berylium · · Score: 4, Insightful

    From the site:

    PILOT STATUS 5/2004
    * The first 100TB Rack is up and running!
    * The second 100TB Rack will be up by the end of May
    * Thermal Targets have been met
    * Systems Booted from USB Dongle
    * Reiser FS running
    * PC-based Router running


    Maybe I'm missing something but this looks to me like they don't really have a Petabyte of storage working but plans to incorporate a Petabyte of storage with only 100 TB up and running now. Not that 100 TB is anything to brush off.

    1. Re:Not really a Petabyte...yet by Anonymous Coward · · Score: 0

      Unless anybody actually needs the full capacity *right now*, there's absolutely no point building it. With the quantity of hardware they use and with the always-falling cost of that hardware, putting off building out the capacity a couple of months could allow them to save huge amounts of money. And given the expense and possibly customisable features, projects of such scale are always build on demand. They're not going to be mass produced and all they need is a proof of concept.

      And in a way, it's almost like the supercomputer projects that are sometimes better off doing nothing: If someone is looking to build a new supercomputer for, say, 10 years of calculations, they're often better off doing nothing and saving their money for the first eight years. Then they buy a 5x faster supercomputer for cheaper (that can do 10 year's worth of the old computer's calculations in 2 years) - and when the project's over, they've got a much better supercomputer lying around for the next project.

      Anyway, my point is that they've presumably demonstrated that their architecture is scalable and works well for this section. Should anybody actually want to order one, they could build a functioning petabyte array.

    2. Re:Not really a Petabyte...yet by joebeone · · Score: 1

      as I posted above... their news site is out of date. I saw this thing and they have one 2-rack, 250 TB box up and running... which was the big question if they could get that running.

  39. one petabyte? by circletimessquare · · Score: 4, Funny

    one petabyte ought to be good enough for anybody

    --
    intellectual property law is philosophically incoherent. it is your moral duty to ignore it or sabotage it
    1. Re:one petabyte? by BrynM · · Score: 0, Offtopic
      Even though I'm throwing away some Karma, I can't resist responding to your sig...
      C:\>tracert life.liberty.pursuit-of-happiness
      "Unable to resolve target system life.liberty.pursuit-of-happiness"

      So I pinged it to make sure it was available. The response:

      "Unknown host life.liberty.pursuit-of-happiness"

      I should have known...

      --
      US Democracy:The best person for the job (among These pre-selected choices...)
    2. Re:one petabyte? by kfg · · Score: 1

      Unless you've got one of those new record your life on video so you can live it vicariously later phones or something.

      Face it, the technology is going to obsolete any amount of storage.

      KFG

    3. Re:one petabyte? by isorox · · Score: 1

      Lets assume that a person can process data at 30GBit per second (10 times the rate of uncompressed top of the line HDTV), that's less then 80 hours of information. In a hundered year life span you'll need 12,000 of these. However, the time is coming where that milestone will be reached. 1GB was a pipe dream to consumers 15 years ago, Now 1TB is only a few hundered bucks. in 50 years we'll be looking at 1EB on the desktop, in 100 years I doubt anyone could digest the information on a consumer hard drive.

      However what about the stuff we don't see. The average paperback book has 10^26 atoms. Say each atom's information (ignore Heisenberg) could be stored (somehow) using 1 bit, you'd need a Trillion of these arrays. Imagine trying to store a copy of the planet on a HDD? There's always a use for large ammounts of storage.

    4. Re:one petabyte? by pyrrhonist · · Score: 1
      "Unknown host life.liberty.pursuit-of-happiness"

      Let's ask the White House about this:

      $ telnet whitehouse.gov
      Trying 63.161.169.137...
      Connection closed by foreign host.
      $
      Huh? Since when does the White House have a foreign host?!?

      --
      Show me on the doll where his noodly appendage touched you.
    5. Re:one petabyte? by Saeger · · Score: 1
      in 100 years I doubt anyone could digest the information on a consumer hard drive.

      In 100 years (actually, more like ~25), I doubt that we'll still be so limited by our slow organic brains. It's not just hard drives that are improving exponentially...

      --

      --
      Power to the Peaceful
    6. Re:one petabyte? by Anonymous Coward · · Score: 0
      Imagine trying to store a copy of the planet on a HDD?

      Wouldn't the act of storing that information cause the information to change? How do you store the data about the HDD?

    7. Re:one petabyte? by Anonymous Coward · · Score: 0

      I would much rather initiate wholescale nuclear war so that we can eradicate naive and arrogant humans like yourself.

      Have you ever heard of hubris?

  40. Hmmmm... by Anonymous Coward · · Score: 0

    I guess I'll buy a couple for my porno collection. Hope it's enough room...

  41. "a million gigabytes"... by TrozPoit · · Score: 0
    "one petabyte of information (a petabyte is a million gigabytes)"

    Isn't a petabyte 1,048,576GB? I'm not being pedantic here... that's an extra 48,576GB, afterall.

    1. Re:"a million gigabytes"... by I+confirm+I'm+not+a · · Score: 2, Informative

      As always, wikipedia has the answer(s):

      Because of irregularities in definition and usage of the kilobyte, the exact number could be any one of the following:
      1. 1,125,899,906,842,624 bytes - 10245, or 250. This is 1024 times a terabyte. This is the definition used in computer science and computer programming
      2. 1,000,000,000,000,000 bytes - or 10 15.

      Damn! Ambiguity!

      --
      This is where the serious fun begins.
    2. Re:"a million gigabytes"... by Anonymous Coward · · Score: 0

      No. A petabyte is 1e15 bytes. If you want to be pedantic about it, then true pedantry must acknowledge that in SI, powers-of-two have their own prefixes. You're thinking of a pebibyte. (1 PiByte, which is 1024 TiByte, or 1048576 GiByte, not GByte.)

      Here is a list of the binary prefixes.

      Memory is the only component with sizes still commonly quoted with abused powers-of-ten prefixes. (That is, using 1 "kilo" byte meaning 1024 bytes, instead of saying one kibibyte or meaning 1000 bytes.) Hard drives have had sizes quoted in powers of ten for at least a decade.

    3. Re:"a million gigabytes"... by timmi · · Score: 1

      realisticlly, Ram units are bytes/2^10 (1024)

      hard drives are strictly 10^3 (1000)

      and if you look an the capacity in bytes of a hard drive, the total space is >= rated size in GB * 1,000,000,000

      Not sure how RAID arrays would affect this, (size of FAT or NTFS Index relative to cluster size...)

    4. Re:"a million gigabytes"... by nacturation · · Score: 1

      Google weighs in on the answer as well.

      --
      Want to improve your Karma? Instead of "Post Anonymously", try the "Post Humously" option.
    5. Re:"a million gigabytes"... by Anonymous Coward · · Score: 0

      how about you start knowing that Kilo and Mega have been used in CS since day 1 and a byte is not an SI unit you fucknut!

  42. And to think.. by Rytr23 · · Score: 1

    I was just thinking to myself the other day.. where am I gonna get storage solution that'll give me not only a petabyte of storage but more importantly, is also shipping container friendly? ..will wonders never cease!? How huge is the market for this type of device? I mean seriously.. How many entities were clamoring for this..and how many could they possibly utilize?

    --
    So many injustices..so little time..
    1. Re:And to think.. by Local+Echo · · Score: 1

      I think it's more of a case of see what I can do rather then doing something useful

    2. Re:And to think.. by TTK+Ciar · · Score: 1


      I'm not at the liberty to say too much, but we do already have a few entities asking for them (either full petaboxes, or fractional petaboxes).

      It turns out that more than a few national libraries have been looking for something like this.

      -- TTK

  43. To bad it won't last... by gremlins · · Score: 5, Insightful

    I know the pull is to get these things as big as you can get but i would love to see hard drives that will work for ever. Now I know everything breaks but I mean in 400 years how is anyone going to know what we were like if all the data on us slowly goes away because the hard drives or the cds don't really last very long

    --
    just because your a schizophrenic doesn't mean people arn't really out to get you
    1. Re:To bad it won't last... by ezzzD55J · · Score: 0, Offtopic

      Please someone mod this man up, because he's right. :)

    2. Re:To bad it won't last... by Tailhook · · Score: 4, Funny

      ...but I mean in 400 years how is anyone going to know what we were like...

      What tiny fraction of our history is actually preserved in a useful manner will be misinterpreted and spun in ways you can't possibly imagine. 400 years from now you will be known as an ignorant fool guilty of untold crimes against lord knows whom. This will be true regardless of the quality of the archive used to figure you out.

      So don't worry about it.

      --
      Maw! Fire up the karma burner!
    3. Re:To bad it won't last... by TTK+Ciar · · Score: 1


      We have a little experience with hard drive failures at the Archive :-)

      The solution we've settled on is to make the data self-repairing, and make the hard drives fast and easy to swap in/out (but without driving up the per-node cost too much .. so no hot-swappable SCSI). So as disks blow up, a sysadmin replaces it with a new one, which gets re-filled with its previously-stored data.

      As you might imagine, our investigations into the lifespans of different models of hard drives is ongoing :-) but I'm not sure of just how much I should say .. if I said "oh yeah, wee-dae-yung's WetTissuePaper-100 really sucks", would wee-dae-yung's lawyers show up and sue me (or the archive!) into oblivion?

      I'm sure the information will get out there somehow, though .. if nothing else, there are anonymous channels .. (whistling an innocent tune)

      -- TTK

  44. Re:wrong by ResidntGeek · · Score: 1

    Wow, that was a bit late... stupid dial-up.

    --
    ResidntGeek
  45. That's cool, but..... by j3ll0 · · Score: 1

    will they also supply a petabyte capable backup solution?

    I ain't changing the tapes on that thing !!!

    1. Re:That's cool, but..... by csirac · · Score: 1

      At 1.1PB per L550 Silo, from Sun Microsystems.

    2. Re:That's cool, but..... by vlag · · Score: 1

      Also available, the Sony PetaSite. A lot cheaper, but slower than the Sun box.

      --
      Do you want to remove linux?
  46. Re:wrong by Anonymous Coward · · Score: 4, Funny

    After extensive market research, it has been determined that a "buttload" is roughly 3.5 gigabytes.

    In case you were wondering.

    It's my favorite unit of measure.

    Coincidentally, I have a buttload of porn.

  47. Re:wrong by Loconut1389 · · Score: 2, Informative

    Depends on whether you're talking Pb or PBi.. If i recall with the big HD size debate several months ago, the Gb/Mb are multiples of 1000 whereas GBi/MBi are multiples of 1024.. maybe i have the abbreviations wrong.. but there are separate units for 1024 multiples due to some whacky issue with SI units or something.. does anybody remember the link to that thread?

  48. Yeah but... by ErichTheWebGuy · · Score: 1

    Does it run Linux? :)

    --
    bash: rtfm: command not found
    1. Re:Yeah but... by Anonymous Coward · · Score: 0

      I speak for everyone when I say "Linux is dying, get over it"

    2. Re:Yeah but... by Anonymous Coward · · Score: 0

      sorry I have to. RTFA!!!

  49. Obligatory refrences: by Anonymous Coward · · Score: 0
    Imagine:

    A beowulf cluster, and all the pr0n!

    There, done.

  50. Information overload / archival quality? by jeephistorian · · Score: 0

    We're so informtaion heavy, will we need things this huge for acrchiving? How will the important stuff be seperated from the chaff (redundant stuff like /. dupes)?

    Fritz

    ______________

    --
    Huh?
  51. Backups: That's a big stack of DVD-R disks by G4from128k · · Score: 1

    My quick calc suggests that a petabyte would require about 213,000 DVD-R disks at 4.7 GB/disk. At about 1.2 mm/disk, that's a stack about 255 m (837 ft )high.

    I don't even want to think about backing this up on a million some-odd CD-Rs. I suspect that the first CD-R would have rotted long before the last CD-R was written.

    --
    Two wrongs don't make a right, but three lefts do.
    1. Re:Backups: That's a big stack of DVD-R disks by Dyolf+Knip · · Score: 1

      Minor nitpick. Hard drives are measured in 1KB=1000 bytes but CDRs are measured in 1KB=1024 bytes. In fact, every 700MB disc I've seen was actually 703MB ~ 737,000,000 bytes. Applying the rest of your calculations... 2826 days. A good 5 months less!

      --
      Dyolf Knip
  52. Aha! by Dark+Lord+Seth · · Score: 3, Funny

    See? Those Longhorn specs are quite easy to achieve... Now let's sit back and wait for Intel or AMD to come up with a 1x1x1m slab of silicon that can melt graphite and run Longhorn at the same time!

    1. Re:Aha! by The+Master+Control+P · · Score: 1

      I would be most interested in a device that can remain coherent under the 8000*F and several hundred bars of pressure needed to make liquid carbon AND perform the 10 quintillion operations per second minimum for Longhorn :)

  53. Useless Statistics! by INMCM · · Score: 5, Funny

    1 MILLION GIGS! BAH! That isn't news unless they convert it to some entirely inappropiate metric. How many Library of Congresses is this? How many 128kbps MP3s can you store on it. And most importantly, how many floppy disks is this equivalent too?!

    --
    Caffeine Good
    1. Re:Useless Statistics! by isorox · · Score: 5, Informative

      How many Library of Congresses is this?

      50

      How many 128kbps MP3s can you store on it.

      250-300 million depending on song length

      And most importantly, how many floppy disks is this equivalent too?!

      700 Million - nearly 40,000 miles when laid end on end, or about 1500 miles when stacked on top of each other.

    2. Re:Useless Statistics! by IntelliTubbie · · Score: 4, Informative

      How many 128kbps MP3s can you store on it.

      Glad you asked. Assuming that we have a 10^15 byte disk (which is how those decimal-loving hard disk manufacturers would define it), and your MP3s are encoded at 128kbps (where 1 kb = 1024 b = 128 B, as the binary folks would have it), then you could listen to MP3s nonstop for:

      10^15B/(128kb/s * 128B/kb) = 61035156250 seconds ... which works out to just about 1934 years without hearing the same song twice.

      Cheers,
      IT

      --

      Power corrupts. PowerPoint corrupts absolutely.

    3. Re:Useless Statistics! by glwtta · · Score: 1

      You forgot Human Genomes - I for one can't tell how much storage something has until it is expressed in Human Genomes.

      --
      sic transit gloria mundi
    4. Re:Useless Statistics! by /dev/trash · · Score: 1

      If i had a petabyte of storage, I sure as hell wouldn't be COMPRESSING my music into OGG or MP3.

    5. Re:Useless Statistics! by reboot246 · · Score: 1

      which works out to just about 1934 years without hearing the same song twice

      It's a shame that there isn't that much good music to listen to for the next 1934 years.

    6. Re:Useless Statistics! by Combuchan · · Score: 3, Informative

      > >How many Library of Congresses is this?

      > 50


      According to this article, a Library of Congress is approximately 10 TB (who knew--this obtuse metric actually has a measurement!!!)

      http://articles.findarticles.com/p/articles/mi_m 0B MD/is_39_9/ai_98189690

      So the device actually can contain 100 Libraries of Congress.

      --
      "[T]he single essential element on which all discoveries will be dependent is human freedom." -- Barry Goldwater
    7. Re:Useless Statistics! by aboyko · · Score: 4, Interesting

      How many Library of Congresses is this?

      50

      Oh, it isn't, either. Will you people knock it off already with the Library of Congress == 20TB comparison? It's some sort of inane computation made as if the collection were only books, and all the books were represented as ASCII text only. Well, guess what? It's not, and they're not.

      American Memory alone is a good bunch of terabytes, and that's just a wee digitized slice, just several million objects, of all the stuff in the Library. There's a lot. Of Stuff. A lot a lot a lot. Pictures. Maps. Movies. Big ol' stuff.

      Well, I feel better. Thanks!

    8. Re:Useless Statistics! by jlanthripp · · Score: 1

      That would be approximately 100 Metric Shitloads, or 128 Imperial Shitloads.

      --
      "Alcohol, Tobacco, & Firearms" should be a convenience store, not a government agency.
    9. Re:Useless Statistics! by carlislematthew · · Score: 3, Funny
      "40,000 miles when laid end on end"

      What the hell does that mean? How many times around the earth is that? How many times to the moon and back?

      START MAKING SENSE!

    10. Re:Useless Statistics! by adpowers · · Score: 1

      I can't wait for Steve Jobs to announce the iPeta. "300 million songs in your closet."

    11. Re:Useless Statistics! by BollocksToThis · · Score: 1

      And most importantly, how many floppy disks is this equivalent too?

      Floppy disk (n): Low cost media designed to store data for x-1 seconds, where x is the time required to get the data to it's intended destination.

      This thing will never match that functionality.

      --
      This sig is part of your complete breakfast.
    12. Re:Useless Statistics! by addaon · · Score: 1

      Too bad it's too late to use that for some return-of-the-lord thingy.

      --

      I've had this sig for three days.
    13. Re:Useless Statistics! by Anonymous Coward · · Score: 0

      And most importantly, how many floppy disks is this equivalent too?!

      call me ignorant, but: _what_ discs?

    14. Re:Useless Statistics! by Quixadhal · · Score: 1
      And most importantly, how many floppy disks is this equivalent too?! 700 Million - nearly 40,000 miles when laid end on end, or about 1500 miles when stacked on top of each other.
      So, what you're saying is that when we get carbon tubes up to orbit, they could make a giant raid array of how many floppies, using the filimant as the spindle?

      That would be about... what... 8PB or something?

  54. I can see the office manager now... by stienman · · Score: 4, Funny

    Office manager: "Hey, Adam, do you know why our power usage might have gone up this last month?"

    I surreptitiously conceal the firewire cable going out the side door.

    Adam: "No, John, I haven't the foggiest."

    OM: "Ok, well I'll ask Kim when I talk to her about the strange shipping container outside. Thanks."

    -Adam

  55. A petabyte here, a petabyte there... by rikomatic · · Score: 2, Funny

    ... and pretty soon you're talking REAL storage.

  56. pedo, not peta by Anonymous Coward · · Score: 1, Funny

    you perverts try to hide your pr0n, but we know what you do with these things...

  57. Immortality? by Sir+Nimrod · · Score: 5, Funny

    In his novel 3001 Arthur C. Clarke asserted/speculated that one petabyte would be sufficient space to store a lifetime's memories. (He didn't say if this was compressed.)

    So, assuming you can handle the trivial exercise of transferring your memories (the implementation of which is left as an exercise for the reader), immortality is yours for the buying!

    1. Transfer memories to Petabox. Sign with your public key, so everyone knows it's you. Don't encrypt!
    2. Put Petabox in shipping container, along with retrieval instructions in English, Esperanto, and Chinese (to cover your bases).
    3. Bury shipping container in Yucca Mountain. (It's unlikely to ever see any nuclear waste, and it'd be a shame to waste the space.)
    4. Kill yourself.
    5. Wait for a society (a) advanced enough to restore you and (b) rich enough to bother.
    --
    The United States of America: We mean well.
    1. Re:Immortality? by Marxist+Hacker+42 · · Score: 0

      IIRC- in that novel the main use of the "petabyte chip" (a petabyte of flash memory on some storage device ~= the size of a credit card) was in full motion/full sound cameras that people wore as lapel pins. It's a simple matter from there to time/date stamp the recordings of a full life, for later search & retrieval. Ted

      --
      SJW: a person who perceives an injustice, and while correcting it, commits a greater injustice.
    2. Re:Immortality? by theLOUDroom · · Score: 4, Funny

      Transfer memories to Petabox. Sign with your public key, so everyone knows it's you. Don't encrypt!

      Don't encrypt! Ye gods man!
      I don't know about you, but I have a few things I wouldn't like people to know about, even long after I'm dead.
      I mean, let's face it, not all our memories are that flattering.

      And anyways, I'm pretty sure some of the memories from my college years are already "encrypted".

      --
      Life is too short to proofread.
    3. Re:Immortality? by Onewheel · · Score: 1
      1. Transfer memories to Petabox. Sign with your public key, so everyone knows it's you. Don't encrypt!
      2. Put Petabox in shipping container, along with retrieval instructions in English, Esperanto, and Chinese (to cover your bases).
      3. Bury shipping container in Yucca Mountain. (It's unlikely to ever see any nuclear waste, and it'd be a shame to waste the space.)
      4. Kill yourself.
      5. Wait for a society (a) advanced enough to restore you and (b) rich enough to bother.
      Hm... I don't see the "6. Make profit!", something is wrong.
    4. Re:Immortality? by LaCosaNostradamus · · Score: 1

      Yucca would be a poor choice. 1) Contamination by radiation would surely degrade the storage media for any reasonable length of cultural-change time. 2) Temperatures in the actual storage areas might get as high as 500 degress F from the heat generated by the materials inside the waste casks. That would also threaten your storage media.

      --
      [You have a stable society when some nut guns down a schoolyard and the law doesn't change.]
    5. Re:Immortality? by Compulawyer · · Score: 1
      Actually, I believe it is:

      6. ???

      7. PROFIT!

      --

      Laws affecting technology will always be bad until enough techies become lawyers.

  58. Re:wrong by Anonymous Coward · · Score: 0

    I have a question, I dont know many americans in real life, and have never had the chance to ask this, were the back to the future films misleading me when they taught me that Yanks pronounce the Giga prefix "Jigger", because if you really walk around saying that to each other in the office, that would rock.

  59. WoW by nevek · · Score: 1

    Isint it great that in 5 years that will probably all fit onto a disc slightly bigger than a penny.

    Hey, at least I didnt say (so much pr0n)

    1. Re:WoW by TTK+Ciar · · Score: 1


      Unlikely to be quite that soon .. there are 3200 disk drives in a completed petabox system. It would take about 12 doublings of disk drive density to put that much storage into a single drive.

      Just before the dot-bomb went boom, hard drive storage was doubling at about 2.1x per year. A few years ago it was about 1.6x per year. So if we naively assume that improvements will continue as a geometric function of time, figure it will take between 12 and 18 years for a petabyte-sized hard disk drive to come to market.

      -- TTK

    2. Re:WoW by Dyolf+Knip · · Score: 1

      Do you realize how disgusted that makes me feel, as I sit here at home and struggle to get this 8-drive 1TB RAID5 array running right?

      --
      Dyolf Knip
  60. Degaussing Cannon Ready by VanWEric · · Score: 1

    Imagine hitting that with a nice heavy dose of degauss. The resulting random data could possibly be a functioning program... Could we create computer life? You'd need to try and find these programs to get them started, but I bet you could create digital life with this puppy.

    --
    www.olin.edu
    1. Re:Degaussing Cannon Ready by HermesHuang · · Score: 1

      Question is how do you deliniate when one "program" ends and another begins? At least when they zap primordial ooze to try to create life the individual "organisms" are clearly separated. And would the program be in assembly? Bytecode? Some programming language yet to be invented? There is unfortunately too many variables for such an experiment to work very well. At most you could try to scan for "begin" and "end" type commands and see if the stuff in between would also execute. On the other hand, if someone does find a clever way to do the experiment, Monte-Carlo calculations would have just gotten a huge boost. And national labs would be buying huge hard drives instead of supercomputers.

    2. Re:Degaussing Cannon Ready by waferhead · · Score: 1

      Might as well suck from /dev/random and run the result thru a genetic algorithm framework, like genesys, and let the fittest survive.

      (wait a minute, that's how it works...)

  61. The cost... by brokenspark · · Score: 2, Informative

    "Rack materials cost is currently estimated to be $121K for 96TB"

    http://www.archive.org/iathreads/post-view.php?id= 13509

  62. read the docs by curator_thew · · Score: 2, Interesting

    If you look down in the message list, you see a reference to pdf + ppt docs. Here's another related project Planet Ten Modular Data Centers.

    Yes, it's a petabyte once you fill the shipping container. Honestly, I thought of this idea last year (using stock shipping containers), and now I'm fascinated that they've made it happen.

    My only suggestion is that this is prototype: the eventual production systems (say, a couple of years time) should have custom shipping containers for:
    * any of the side panels can open to access a rack and hot swap failing racks, so there is no need for a middle entry aisle
    * the cooling system should be built into the structure, like existing refigerant containers
    * not just data storage, but also computing facilities

    1. Re:read the docs by TTK+Ciar · · Score: 2, Interesting

      > * the cooling system should be built into the structure, like
      > existing refigerant containers

      There's been discussion/research at the Archive regarding exactly this :-)

      > * not just data storage, but also computing facilities

      Each petabox node is a completely functional ITX PC, running Debian Linux, so it is already also a computing cluster. It's not a very good one, unfortunately, because in the interest of keeping the heat and power down a very low-power processor was used, an 800MHz Via C3, which is marginal as an integer-intensive processor.

      Also, the storage nodes (which comprise 796 of the 800 nodes) only have 100bT ethernet, which would limit the petabox's ability to efficiently make use of what processing power it does have (in the data distribution phase).

      Nonetheless, 796 800MHz processors is nothing to sneeze at. When the PetaBox is being a PetaBox, its processors are mostly idle. I'm sure folks will figure out something to do with those spare cycles (or perhaps not, considering that exercising the processors noticeably increases the power and cooling burden).

      -- TTK

    2. Re:read the docs by curator_thew · · Score: 1


      Nice reply. Certainly this is at any early stage, and it will be interesting to see later production refinements. I notice BBC archivist commenting about "selective powering": a very interesting comment.

      In general, it would be interesting to build a business around the whole concept of "computing in containers": for example, some containers that are part hard drive, and part juke box (perhaps too expensive and unreliable though ...).

  63. well i know that... by ambienceman · · Score: 1, Funny

    This is totally unnecessary. EVERYONE knows we only need 640k

  64. Hmm by XMyth · · Score: 1

    This is just in time for me to start my HD porn archive!

  65. Re:wrong by isorox · · Score: 1

    Specifically, 1,048,576 GB

    No, Specifically, 1,000,000 GB. GB == Gigabyte. Giga == 10^9, always. No "in computing it doesn't" crap. Giga is an SI prefix.

  66. But not Peta by Dog135 · · Score: 1

    Yeah, they did, but that was the gigabyte model. This is the petabyte model now.

    It's all just symantics. Nothing new.

    --
    "That's so plausible, I can't believe it!" - Leela
  67. Re:wrong by Anonymous Coward · · Score: 0

    Have any of you nerds thought that maybe YHBT?

  68. In other news... by nacturation · · Score: 2, Funny

    Duke Nukem Forever is "ready"!

    --
    Want to improve your Karma? Instead of "Post Anonymously", try the "Post Humously" option.
    1. Re:In other news... by rozz · · Score: 0

      and in OTHER news Phantom console is also "ready" ... and that for the SECOND time!

      --
      "There is nothing more frightful than ignorance in action." Johann Wolfgang von Goethe
  69. Porn-o-rama by Anonymous Coward · · Score: 0

    That's a lot of porn in one space.

  70. Comment removed by account_deleted · · Score: 1

    Comment removed based on user account deletion

  71. Hard drive lifespans by MikeBabcock · · Score: 2, Interesting

    I'd love to see an equivalent to all the benchmarking websites out there for telling me what hardware is reliable, and not just fast. I already know what the fastest drives, fastest video card, quietest fans, etc. are, but which ones last longest? Which drives *never* have failures that affect real data? Which cables are properly certified and insulated for high-volume transfer in a confined space rubbing up against other cables? Etc.

    If you know of such a site, tell me.

    --
    - Michael T. Babcock (Yes, I blog)
    1. Re:Hard drive lifespans by Zordak · · Score: 1

      Get into nuclear weapons. There are people who do that for a living.

      --

      Today's Sesame Street was brought to you by the number e.
    2. Re:Hard drive lifespans by pair-a-noyd · · Score: 1

      I have hard drives over 20 years old that I can slip in a machine and STILL power up and retrieve the data, 100% intact.

      I have CDR's less than 2 years old that are SCREWED beyond hope because of CD rot..

      After being bitten by the CD rot bug I have decided to buy 2tb worth of hard drives that I will use to dump my CDR's into, unplug and store them in a fireproof safe.. 20 years from now, my data will still be there. I will NOT be depending on CD or DVD media for backup anymore..

    3. Re:Hard drive lifespans by jonman_d · · Score: 1

      Pen. Paper.

      The truth is, it's going to be a damn long time before we develop ANY computer hardware that'll stand the test of time. It's all just too fragile. If you really need something to be stored forever, print it out. They make special archival-quality paper and inks that are designed specifically to last hundreds of years.

    4. Re:Hard drive lifespans by Nasarius · · Score: 1

      Your hard drive will last forever too, if you fill it up then don't use it.

      --
      LOAD "SIG",8,1
    5. Re:Hard drive lifespans by Have+Blue · · Score: 1

      I don' think it would be too hard to design a medium which could last essentially forever (MO disks?), but the catch is that you also have to have an operational media reader, which is a far more complex and delicate item. At the very least you could preserve the design and specification for the media reader on paper, but then reading the media becomes a very expensive operation if you have to rebuild the reader from scratch.

      Hard disks don't fail because something happens to the magnetic disks, they fail from other points but often destroy the physical disks in the process (head crash).

  72. Comment removed by account_deleted · · Score: 1

    Comment removed based on user account deletion

  73. Bah. Just hope it doesn't crash by Anonymous Coward · · Score: 0

    One of the BIGGEST problems with current filesystem technology is that they *all* do a very poor job when it comes to crash recovery, when you start talking over a terabyte of storage.

    Someone PLEASE correct me if I'm wrong, but all the UNIX/Linux filesystems still just merrily dump things in lost+found, with NO organization of the files.

    I've seen, litterally, over 4 Million files in /lost+found alone, with another 8 million in various directories under that.

    Things get *real* ugly when you try to recover from that. find, tar, cpio, ls and other normail utilities ALL break down when you are dealing with a situation like this.

    There needs to be a better system. Current UNIX and Linux technology just can't handle it.

    So if you move into this space, the only viable solution right now is to make certain you have excellent backups, which is another issue in itself

    1. Re:Bah. Just hope it doesn't crash by TTK+Ciar · · Score: 1


      Fortunately, you are mistaken. The PetaBox employs ReiserFS (which would not have been my first choice, but I will work to any specification given to me), which is a fully journaled filesystem. It does very well recovering from unscheduled downtimes.

      ReiserFS has thusfar shown itself a robust solution to this problem, but so would Ext3, and probably XFS (which I know little about, but Joerg here at the Archive likes a lot). You are thinking of Ext2.

      -- TTK

  74. Is it me, or does "Petabyte"... by MultisSanguinisFluit · · Score: 1

    sound like a technical 'zine for animal rights advocates?

    --
    > get tea
    No Tea: dropped.
  75. Re:wrong by ResidntGeek · · Score: 1

    The abbreviations are MiB, GiB, and PiB.

    --
    ResidntGeek
  76. Microsoft by HermanAB · · Score: 1

    will find a way to fill it up... ;-)

    --
    Oh well, what the hell...
  77. Actual Capacity Less... by Meneudo · · Score: 1

    Of course, you have to figure in that by Petabyte, they mean 1,000,000,000,000,000 bytes. Of couse, the actual formatted capacity is much less.

    1,000,000,000,000,000/(1024^4) yields only about 909.5 TB. Correct me if I am mistaken.

    --
    ...
    1. Re:Actual Capacity Less... by TTK+Ciar · · Score: 1


      PetaBox v1.0 would have a formatted capacity of 937.4732544 * 10**15 bytes, but the first PetaBox is not going to be 10 PetaBox-v1.0 racks. It will probably be either 1 or 2 v1.0 racks and 8 or 9 v1.1 racks.

      I cannot say for certain, but it is my understanding that PetaBox v1.1 will use slightly different models of hard drives than the ones used in this rack here. It would have a formatted capacity of about 1094 * 10**15 bytes.

      (We've been putting this first rack through its paces, and have learned a lot. v1.1 will reflect that experience with a few minor tweaks.)

      -- TTK

    2. Re:Actual Capacity Less... by gumbi+west · · Score: 1

      Actually a Petabyte is 1x10^15 bytes, but it is only 909 Tebibytes.

    3. Re:Actual Capacity Less... by Anonymous Coward · · Score: 0

      Is that not what parent just said?

    4. Re:Actual Capacity Less... by gumbi+west · · Score: 1

      Sorry, the point is that there is a special way of reffering to the 2^10 type powers, and it isn't peta or tera. NIST spells this out at the linked pages.

  78. Interesting link... by Noryungi · · Score: 4, Interesting

    It's right there under the pictures :

    http://capricorn-tech.com/

    The site is rather empty right now, but it seems this is the company that will market this petabyte machine... er... box... er... whatever the name is.

    --
    The right to offend is far more important than the right not to be offended. (Rowan Atkinson)
    1. Re:Interesting link... by pair-a-noyd · · Score: 1

      Bad show old man, bad show...
      I see your MAC address there..

      Ah, at long last, my very own petabyte storage....

      Life is sweet....

  79. themselves, not google... by SuperBanana · · Score: 3, Funny
    Can we say, Goooooooooooooooogle?

    I'm guessing they were referring to themselves, not google.

  80. Imagine a Beowulf... by big+daddy+kane · · Score: 0, Troll

    sorry but it's kinda required.

  81. Re:wrong by joe_bruin · · Score: 1

    ah, you seem confused. bytes are not SI units, those would be octets (an octet is 8 bits). a kilo-octet is 1000 octets, and a peta-octet is equal to 1,000,000 giga-octets.

    a byte is from the computing world. it's a computing term. it was invented for computers. before people were sending bits over networks and before hard-drives existed, a kilobyte had been defined as 1024 bytes. that's how it works in the computer world.

  82. Comment removed by account_deleted · · Score: 2, Interesting

    Comment removed based on user account deletion

  83. Yeah but.. by Anonymous Coward · · Score: 0

    ..in 10 years what are the odds that you'll lose a [ disk | holographic cube | whatever ] with that much storage capacity down the back of the sofa.

    And you won't care because it'll be less effort to format a new one than hunt for the old one.

  84. Nice design by mfarver · · Score: 1

    Very nice, clean design overall. I'd like to know who the supplier was for the rack cases, or if they had them made custom. I've seen the half cases with the ports on the front before, but I have not been able to locate a US supplier.

    It appears that the nodes are half sized, allowing for 40 systems per side, or 80 systems total. The null modem console cables connecting node pairs together is clever... if any one machine fails, you can restart it as long as its neighbor is still alive.

    Mark

  85. Missing the forest for the trees by magicclams · · Score: 1

    I think we're all jaded. This is a device that's big enough to archive the contents of every library in the world and we're complaining about the fact it won't fit in our front pocket.

    1. Re:Missing the forest for the trees by shigelojoe · · Score: 1

      No, just the fact that it won't fit into a VW bus. I mean, come on, cubic feet? Give me a unit of measurement I can *use*!

  86. Re:Finally - monkeys by Splork · · Score: 2, Interesting

    monkeys eventually write both vi and emacs while working on shakespeare

  87. This really matters to some people by Anonymous Coward · · Score: 0

    Yeah, ok, we're all making fun of this type of product. But, lot's of businesses really care about this kind of thing. Forget google. Think about an airline. They need to store boatloads of information every day- flight arrival/departure info, passenger names, etc, and this information is never deleted.

    You have a situation in which your probably accumulating MBs or GBs of new information every day and you need to store all of it for a long time. A PB of storage is nothing for this type of application. Same deal for target/wallmart -- inventory stats, every transaction at every register, etc. Storage is a big deal for a lot of business-types. Find a way to make storage more compact, more easily backed up, etc and theres some real money to be made.

  88. There's an easier way to do this... by exp(pi*sqrt(163)) · · Score: 2, Insightful

    ...just mount /dev/random as a petabyte drive. Admittedly it might be hard to find your data in there - but chances are it is in there somewhere.

    --
    Doesn't it make you feel good to know that our freedoms are protected by politicans, lawyers and journalists.
  89. Re:wrong by Elwood+P+Dowd · · Score: 4, Interesting

    Everyone I know says giga with a hard G. The only exception I know of is Christopher Lloyd's character in Back To The Future.

    --

    There are no trails. There are no trees out here.
  90. Re:wrong by Anonymous Coward · · Score: 0

    how many VW Bug kilobits/Library of Congress Seconds does this thing get anyway

  91. Makes life easier by esvoboda · · Score: 2, Funny

    Good, now I don't need to delete my spam.

  92. to quote a movie by Anonymous Coward · · Score: 0

    yeah, but does it play dvds?

  93. actually... by MarcoAtWork · · Score: 4, Funny

    given that monkeys have basically four hands I think they'd be more suited for emacs (alt-meta-control-super-hyper-shift-q) than vi ;-)

    I have been using emacs for nearly 10 years now and I swear sometimes I have been seriously considering adding a foot pedal or 2 to my setup (besides control, shift and meta I also routinely use Super and Hyper, xmodmap is great!)

    --
    -- the cake is a lie
  94. Re:wrong by isorox · · Score: 1

    Irellevent, kilo was arround, and meaning 10^3, a long time before computers. The IEC decided to use kibi 6 years ago, although kibi isn't an SI prefix. You can't use 1 word for two different meanings in science. 1 slug-pound-foot per bushel isn't apropiate anymore.

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

    80 systems, 4 drives system = 320 drives.
    Copan puts 896 drives in one system
    http://www.copansys.com/products/specifica tions.ht m

  96. it just has to be said. by l33t-gu3lph1t3 · · Score: 1

    oh man, imagine a beowulf cluster of those...;)

    --
    ------- "From bored to fanboy in 3.8 asian girls" ----------
    1. Re:it just has to be said. by hakr89 · · Score: 1

      i'm not sure...but i think you can turn it into a beowulf cluster

  97. Re:wrong by isorox · · Score: 1

    We need 1.21 jiggawatts!

    (Of course that's only for the 0.1 seconds before the car went back to the future - the ammount of power it used was arround 35kWh, so you could charge it up using a 13A/240V mains overnight.

  98. Oh please! by mog007 · · Score: 1

    Gimmie the yottabyte drive already!

    A trillion Terrabytes!
    A quadrillion Gigabytes!

  99. Re:wrong by Marxist+Hacker+42 · · Score: 0

    PiB, GiB, MiB (and also KiB) are all 1000. PB, GB, MB, and KB are 1024 due to BINARY! How much of a non-Geek do you have to be not to realize that 1000 isn't a power of 2, while 1024 is?

    --
    SJW: a person who perceives an injustice, and while correcting it, commits a greater injustice.
  100. Re:Potential customers - a TLA - NSA by Anonymous Coward · · Score: 0

    A Three Letter Agency? Maybe with first letter N and last letter A?

  101. Re:wrong by Anonymous Coward · · Score: 0
    After extensive market research, it has been determined that a "buttload" is roughly 3.5 gigabytes.

    It's my favorite unit of measure.

    I use SI, so I have a metric buttload of porn.

  102. That reminds me... by Daverd · · Score: 1

    of when I first started playing nethack.

  103. Scientific Data by Scott+Ransom · · Score: 4, Interesting

    I'm heavily involved in a 5-6 year project to use the Arecibo telescope to search for new pulsars. The project uses a new 7-beam receiver system, each of which takes data from up to 1024 nearby frequency channels. The data is 16-bit sampled over 15000 times per second from each frequency channel. We need the time and frequency resolution to find exotic millisecond pulsars.

    Over the couse of the survey we expect to take about 1 PB of data. We're still trying to figure out exactly how we will process and store it all.

    For more info, you can poke around here.

    1. Re:Scientific Data by Bryan+K.+Feir · · Score: 1

      Hey, I was one of the people working on the S2 data recorder for radio astronomy work. We actually did work on a computer network storage version of that system, which could have got up to about half a petabyte of data if all tricked out.

      The S3 data recording system would, in theory, have been able to record at one gigabit per second, which is almost what you need here. (7*1024*16*15000 is close to 1.7 Gb/s, by my calculations). We were working on having it run unattended observations for 24 hours straight at that rate by using a robot to change tapes.

      Too bad the S3 system never got past the preliminary prototyping work because of funding issues... which weren't helped by the development of the Mark V disk-based system.

      -- Bryan Feir

  104. Re:wrong by Loconut1389 · · Score: 1

    I understood the difference (due to binary), but mixed up which one was which and reversed the order (MBi as opposed to MiB). And the SI unit versions are the 1000 multiples, the computer user standard are the 1024 multiples. Thats all i meant.

  105. Re:wrong by Anonymous Coward · · Score: 0

    So people who download lots of porn are known as butt pirates?

  106. Bad Math by CustomFort · · Score: 1

    Actually, it would take more then 925 days :

    60sec * 60 sec/min * 60 min/hour * 24 hour/day = 86400

    1 PetaByte = 8*10^7 100 MegaBits

    80000000 seconds / 86400 seconds/Day = 925.9259259259259 Days

  107. HUGE by anethema · · Score: 1

    Shipping container friendly ?
    (20' x 8' x 8')

    Thats fucking huge. Thats the size of a medium bedroom. Maybe a small bedroom, stilll very big.

    Where do they get shipping containers this size?

    --


    It's easier to fight for one's principles than to live up to them.
    1. Re:HUGE by jlanthripp · · Score: 1

      Not surprisingly, standard general-purpose dry cargo shipping containers (20 or 40 feet x 8 feet x 8.5 feet external dimensions) are used in the shipping industry. They're filled with goods, put on container ships, and taken to various ports around the world. Their size also allows two 20-foot containers or one 40-foot container to fit on a standard 48' flatbed trailer, with room for tie-down straps and such.

      --
      "Alcohol, Tobacco, & Firearms" should be a convenience store, not a government agency.
    2. Re:HUGE by cr0sh · · Score: 1

      And used, they are very cheap! I have seen such containers (some with side loading doors!) go for $3000.00 US - that may not be a lot of square footage, but remember this is pretty hefty steel we are talking about - underground bunker workshops are a possibility with these things. Easily modifiable with little work (just a welder!). Modular and stackable - cover with some insulation (a crapload of expandable foam?), add some A/C - cheap and very durable living space...

      --
      Reason is the Path to God - Anon
    3. Re:HUGE by jlanthripp · · Score: 1

      Thanks for the idea! I'll keep that in mind when I become an evil overlord and require an underground bunker complex ;-)

      --
      "Alcohol, Tobacco, & Firearms" should be a convenience store, not a government agency.
  108. doh by pnrgi · · Score: 1

    you know were going to read about a CEO coming up to one of these, and telling one of his buddys, "Hey, you wanna see a hot swap!?"

  109. Scientific Data by madaxe42 · · Score: 1

    These will most likely be primarily used by academic institutions, such as CERN, NASA, and universities who are lucky enough to own particle accelerators. CERN has several petabytes of storage currently, which are housed in silos full of tapes, with rather funky robots which fly around at speed. An average particle collision (single experiment!) will give at least 1GB of data, which will take a considerable amount of time to process into something more managable, and therefore needs storing.
    All in all, I think they'll have a few interested clients, but at the same time, most places where they have minds capable of building machines of the magnitude used for these experiments, they tend to be more than capable of building solutions similar to this.

    Take a look at this!

  110. kazaa by Anonymous Coward · · Score: 0

    download all of kazaa and host your own version of kazaa

  111. WTF? by imag0 · · Score: 4, Funny

    If you gave me a 100 mbit line, it would take me over 92 days to fill it up with porn. More if I slept.

    yeah, but if you looked closer, it's the same 6 gigs over and over again.

    1. Re:WTF? by kabocox · · Score: 1

      If you gave me a 100 mbit line, it would take me over 92 days to fill it up with porn. More if I slept.

      yeah, but if you looked closer, it's the same 6 gigs over and over again.


      I'd be more impressed if you used high resolution cameras and filled it up with your personnally made p0rn myself. Its cheating to use media made by others. You should do it youself. Then wait for it.. open source it.

  112. Petabox? BAH - GMAIL! by Compulawyer · · Score: 4, Funny

    I'm just gonna get 1,000,000 free Gmail addresses and email all my data to myself 1 Gb at a time.

    --

    Laws affecting technology will always be bad until enough techies become lawyers.

  113. Re:wrong by neurojab · · Score: 1

    >Everyone I know says giga with a hard G.

    Giga- derives from the same root as "giant"... One would suspect that's why it's pronounced similarly (or many would claim). Personally, I only started hearing the hard "g" when computers with gigabyte hard drives reached the masses. These same masses also claimed that 3.5 inch floppy disks were "hard disks" because they were less floppy than the 5.25" variety :)

    Gigabyte sounds much cooler with the soft G (Jiga). Let's go with that.

  114. Mmmm.. by nonetheless · · Score: 1
  115. Monkey simulator by bsd4me · · Score: 1

    How about the doctor analyzing zippy?

    --

    (S(SKK)(SKK))(S(SKK)(SKK))

  116. Re:wrong by pchan- · · Score: 1

    You can't use 1 word for two different meanings in science.

    what about "splice"? it has two (opposite) meanings.

  117. No, your math is bad! by Anonymous Coward · · Score: 0

    What kind of math are they teaching the kiddies in school these days? You got confused between m=milli and M=mega.

    He said "100 mbit." I assume he meant that in bits per second. 100 mbit/sec is 100 millibits per second. That's 0.1 bits per second. Your numbers are off by a factor of a billion!

  118. If you're a big fan of this much storage? by stox · · Score: 4, Funny

    Does that make you a petaphile?

    [massive karma burn detected]

    --
    "To those who are overly cautious, everything is impossible. "
  119. Just imagine... by Anonymous Coward · · Score: 0

    a Beowulf cluster of... ... what? Oh.

    In Soviet Russia, old joke redundants you.

  120. booting off a USB dongle by NynexNinja · · Score: 1

    Their site claims they are booting off a USB flash dongle key. This sounds neat. I wonder what the boot time is.

  121. Almost big enough to store the U.S. tax code by rocker_wannabe · · Score: 1

    Including all revisions, of course.

    "You stink therefore you are" - Descarte's mom
    --
    "Meaningless!, Meaningless!" says the Teacher. "Utterly meaningless!"
  122. Re:Petabyte by slickwillie · · Score: 1

    I think that's the name of the Pitbull down the street.

  123. Re:wrong by pair-a-noyd · · Score: 1

    While a "buttload" is a valid unit of measurement and it's good to be aware of it, especially in refering to downloaded pr0n, *MY* favorite unit of measurement is the "red cunt hair". It comes in handy when trying to fit hardware into tight spots.

  124. Re:wrong by Anonymous Coward · · Score: 0
    Giga == 10^9, always. No "in computing it doesn't" crap. Giga is an SI prefix.

    Screw the French.

  125. Livin' on the edge by mikis · · Score: 1

    Apparently they are using VIA chipsets and CPUs and ReiserFS. Hope they have good backup ;)

  126. Ozymandias by Boglin · · Score: 3, Insightful
    Okay, I've heard this too many times and I'm just starting to get sick of it. It's not the computers that are killing your quest for digital immortality; it's just the way that history works.

    You're complaining that these hard drives won't run forever and you're right. Neither will CD's. However, I would also like to point out that the vast majority of ancient egyptian papyrus isn't around today. Also, don't start goign off on using clay or stone tablets, because they break (even the Rosetta stone is broken).

    Honestly, computers are still far superior to what we were using before. It's not like we've got Homer's original version of the Illiad sitting in a museum somewhere; we just have many duplicated copies that have been reproduced over the years. You're right that hard drives fail and CDs break, but we can keep updating onto new media. Besides, when a monk drops an iota when transcribing the Bible, Jesus goes from being God to godlike. When a computer adds an iota, the checkbit fails and the data is resent.

    Somebody is also going to point out that, as systems change, data can become unreadable. Heck, I had a professor who couldn't update his lab instructions because the software that read the lab printouts wouldn't run on new machines and the fileformat wasn't understood by any other software. So, want to stop our data from becoming unreadable? Well, let's just do what the Etruscans did! Of course, we don't have a clue what they did because nobody can read Etruscan. For a more familiar example, think of heiroglyphics before the Rosetta stone. It's pretty common for data to become lost and unreadable. Also, this bring us back to the solution. Along with the data, include the source code for the software that can read it. If you really want to be anal, you could even include the source to an emulator for the machien it was designed to run on.

    Still, you might point out, 400 years from now, we'll still lose 99% of that do to failures of whatever nature. Once again, you would be be right. However, do you honestly believe that we have 1% of all the data that was collected in 1604? Hell, most of the people couldn't even right, so we don't know ANYTHING about their lives. I'm sorry that we can't digitally preserve our wonderous society for all of eternity, but it's completely blind to believe that this makes us in ANY way different to any other culture. Read Percy Shelley's Ozymandias before complaining about how people in the future won't know what our lives were like.

    1. Re:Ozymandias by Suidae · · Score: 1

      All true. The interesting part is that this is probably a good thing for the most part. We generate lots and lots of data that nobody is going to be interested in. The important stuff will probably be preserved, and enough of the cultural/historical stuff will stick around for future historians to figure things out.

      Imagine trying to find anything in 1000 years if every piece of data were saved.

    2. Re:Ozymandias by wizrd_nml · · Score: 1

      I believe the point is that we should be sufficiently more advanced now than Homer or the Egyptians 3000 years ago, that we can find a way to store data for more than a decade (both physically and in a way that it can be understood 100 years from now).

  127. Only on /. by harmonica · · Score: 1

    Yeah and maybe we could clone a few of them (a couple scary and sporty spices) for our own conveniences.....hehe.for singing of course

    Only on /. a comment like this gets moderated as Interesting instead of the non-creepy Funny. ;-)

  128. Format? by eekrano · · Score: 1

    Hopefully it comes pre-formatted in whatever FS the customer wants, otherwise can someone tell me how long this will take to do a full format? hehe.

    --
    -- Eekrano
    1. Re:Format? by Anonymous Coward · · Score: 0

      Since the storage consists of 798 storage nodes, each an independent Linux system with four hard drives per storage node, they can all be formatting their drives in parallel. A given node takes about twelve minutes to format all of its filesystems.

      When a node boots up, its BIOS boot order walks it through the process:
      * Check for bootable USB device plugged into the USB bus. When booted off the dongle, the sysadmin ssh's into the node and runs the NODE.systemimager script, which formats the drive, installs the OS, runs LILO, etc.
      * Check for a PXE bootserver. If a bootserver is found on the local network (the petabox does not have a bootserver by default), it boots up and runs the NODE.systemimager script remotely.
      * Boot off local hard drive.

      The USB dongle has FeatherLinux installed on it. The NODE.systemimager creates seven filesystems and a swap partition:
      * /dev/hda1 / reiser 2GB ............. OS and /root
      * /dev/hda2 /var reiser 1GB .......... logs
      * /dev/hda3 swap swap 1.5GB .......... memory swapspace
      * /dev/hda5 /tmp reiser 148GB ........ scratch space, also backup directories for rsync
      * /dev/hda6 /hda6 reiser 148GB ........ data storage
      * /dev/hdb1 /hdb1 reiser 300GB ........ data storage
      * /dev/hdc1 /hdc1 reiser 300GB ........ data storage
      * /dev/hdd1 /hdd1 reiser 300GB ........ data storage

      If the /tmp directory seems huge, it's because the Internet Archive has a tradition of using it to store a lot of different kinds of applications' data. That convention is expected to continue into ongoing PetaBox software development. Also, rsync uses this space to store backup copies of modified or deleted files (to guard against human error), which could amount to a lot of data.

      -- TTK

  129. Replacing HDs could be a pain by gumbi+west · · Score: 3, Insightful

    If you expect a hard drive to fail after three years (I'm guessing) but these occurances are randomly distributed (an assumption that will be true after running this thing for a year or two) you can then expect that the 4000 hard drives in this array would have about 3 failures per day. This thing would never be at full speed! it would be constantly restructuring its RAID. Also, it would cost about $300 just in hard drives (not to mention controllers, power supplies, et cetera).

    1. Re:Replacing HDs could be a pain by vidarh · · Score: 2, Insightful
      Yes, but you aren't seriously suggesting it would be one RAID over all the disks are you?

      So assuming 3 failures a day, at most 3 RAID's would be running slower a day. Assuming 4 disks per RAID that's at most 12 disks at reduced performance, or 0.3% of the total data set that isn't available at full speed. If that is an issue, you duplicate any data that MUST be available on multiple nodes.

  130. Re:wrong by Anonymous Coward · · Score: 0

    After extensive market research, it has been determined that a "buttload" is roughly 3.5 gigabytes.

    Would that be a metric buttload or an imperial buttload?

  131. Cheap high end storage by minion · · Score: 1

    People keep talking about storage clusters done with PC's using 3Ware or other multi-channel SATA RAID cards. I've got nothing against 3Ware - They run several of my "low end" servers at our office, but what we really love are Infortrend boxes.

    You're not running a full blown OS on each box, the box is a self contained RAID system. We've deployed several of these boxes to clients, typically using the SCSI-to-SATA boxes (shows as a large SCSI device to the HBA, but is running 16 250G SATA drives (4TB RAW!)).

    Its a very inexpensive way to store massive amounts of data and have it directly attached.

    We've just started to look at moving to Fibre. They have a Fibre-to-SATA JBOD getting ready to ship, and with the Fibre head unit, we could expand quite a bit. We've got 12T online now, and are growing at a rate of 2.2G per week. With traditional SCSI based SANs, that would be impossibly expensive.

    --

    -- If we don't stand up for our rights, now, there will be no right to stand up for them later.
    1. Re:Cheap high end storage by Anonymous Coward · · Score: 0

      Dude, this isn't about any of what you are talking about. Did you see the costs?! It's less than $150K/rack which is 100 TB!!!

      http://www.archive.org/iathreads/post-view.php?i d= 13509

      Also, on those drive arrays, you can't run applications directly on top of the disks like you can in a cluster! You have to pull the data off of them, process it, and put it back. I guess you're just talking about a totally different application.

  132. Journaled Filesystems don't solve it. by Anonymous Coward · · Score: 0

    With all due respect, you have absolutely no idea of what you are talking about.

    You've never had to deal with a seriously damaged filesystem. One in which the journal has been lost. Please read the posting again. You totally missed the point.

    ReiserFS in fact is the one which died on me. Sticking everything in lost+found without any intelligence is straight out of the 1970's.

    This is an issue which will bite too many system admins. Only too late will they find their journaled filesystem to end up costing them a lot of downtime because of this stupidity of the current filesystem implementors.

    1. Re:Journaled Filesystems don't solve it. by Anonymous Coward · · Score: 0

      I understand that they aren't a cure-all, but frankly if a filesystem gets so messed up that it is unable to recover, it would probably be best to simply reformat the drive and treat it like a new drive replacing a broken one. Its data would then get restored from the latest backup.
      In the common case, the journal does not get corrupted, and the journalled filesystems can be expected to recover to a well-working state.
      -- TTK

    2. Re:Journaled Filesystems don't solve it. by Anonymous Coward · · Score: 0

      Yes, that's a valid point. But these systems are commonly used for critical servers, where you absolutely need that information now.

      Having a good backup and recovery solution in place is of course desireable. All too many places simply don't understand that. And even decent recovery systems might take days, instead of seconds or minutes.

      It would be real cheap to put lost inodes like 12345 in a directory lost+found/1/2/3/4/12345. Or even lost+found/1/12345 would go a LONG way towards allowing one to quickly recover that critical file immediately. Having millions of files in /lost+found is unacceptable with terabyte and petabyte technology. Something really needs to be done here.

      My apologies for saying you didn't have a clue. Since you didn't address the main point, I had mistaken you for the typical technical dilletant on slashdot.

  133. That's Nice But... by dbretton · · Score: 3, Funny


    Is there enough disk space left to do anything useful after installing WinXP on it?

  134. You think they could spare a boot disk. by Stephen+Samuel · · Score: 1
    Hundreds of Terabytes of disk space, and they're booting off of a friggin dongle????

    I dunno if anybody else laughed at that, but I sure did.

    --
    Free Software: Like love, it grows best when given away.
    1. Re:You think they could spare a boot disk. by vidarh · · Score: 2, Insightful
      Laugh all you want but what they are doing makes sense: Recovering a node in case of a crash or messed up filesystem is easy - you replace the dongle and hit the reset button. No need to have space wasting CD drives or floppy drives, and the rest of the OS can be pulled down over the network.

      The last thing you want with a setup like this is having to haul hardware around or disconnect stuff if you for any reason can't boot of the disks anymore. And you certainly don't want to reduce density by wasting space that could be filled with disks with other stuff.

    2. Re:You think they could spare a boot disk. by Stephen+Samuel · · Score: 1
      I realize that, but it still caught me off guard, and gave me a good laugh. I like life's little ironies, and this is one of them.

      BTW: Dont expect that "the rest of the OS will be pulled down over the network". Remember, this box is likely to be the only fileserver your company is going to need.

      256MB is enough of a filesystem (especially if compressed) to boot up and have all of the needed drivers and networking tools for what could be described as one of the worlds largest embeded systems. Having a 500TB fileserver require part of it's production filesystem be served by yet another fileserver wouldn't be ironic. It would be just plain stupid.

      --
      Free Software: Like love, it grows best when given away.
    3. Re:You think they could spare a boot disk. by Anonymous Coward · · Score: 0

      I should clarify what was meant by "booting off a dongle" :-)

      The 1U cases of the Storage Nodes are as crammed as they can be without compromising airflow. There is no room for a cdrom or floppy drive, but we would like to be able to rescue-boot the Node as needed, and upgrade/install the operating system on a flexible basis.

      At the Internet Archive, the traditional solution to this problem has been network-boot via PXE. This has worked reasonably well, but at the cost of additional complexity.

      To make network-boot reliable, there need to be multiple bootservers, so that if some fail the rest can provide the service. Also, if we need different installations on different Nodes (which we often have), we need a system for managing and targetting different net-boot personalities. We have this, and it works, but it is complicated and difficult to debug when something breaks.

      The learning curve for administering this system is, uhm, steep. The PetaBox is supposed to be manageable by a single system administrator who is no more than human, with little special training. We cannot reasonably expect such a one to debug the net-boot system when it breaks, so we decided to expand our options a little.

      The Nodes are still capable of net-booting, so that we can use them here at the Archive with our current setup, and if a particularly studly PetaBox sysadmin wants to implement net-boot, they can. But the Nodes' BIOSes first check to see if there is a bootable USB-HDD plugged into the bus, and try to boot from that.

      The USB storage devices are 128MB in size (of which we are using about 80MB), and have FeatherLinux installed (thanks, Eric!), along with a script called NODE.systemimager. Once the Node has been booted to FeatherLinux, the sysadmin can log in from the console or remotely via ssh and run the NODE.systemimager script (or do rescue-type stuff). The script formats the partitions, rsync's the OS to /dev/hda1, and runs LILO to make /dev/hda1 bootable. Thereafter, the Node boots from local disk, not from the dongle.

      It is thought that USB-boot will be easier for the average joe to understand, use, and debug -- it's not terribly different from booting off a rescue floppy.

      The NODE.systemimager rsync's the OS from the Homeserver Node, which is an integral part of the PetaBox. It exports an rsync module which contains OSes for homeserver-1.0 and node-1.0, and in the future will contain additional OS "personalities".

      Since the Nodes boot from their local drive, alternate "personalities" can be created by simply modifying a node as needed, rsync'ing it into a new homeserver::petabox/[personalityname] entry, and then rsync'ing other nodes into the new personality.

      A concern is that Nodes' operating system configurations will diverge from the baseline OS image, turning into an administrative nightmare. So we developed an "fsdiff" utility suite, by which an administrator can see at a glance which Nodes have diverged, and how much, and in what ways (ie, which files have changed). This system has been put through its paces as we developed the Node operating environments, and it seems to provide the necessary functionality. It still has room for improvement, of course, but it's all pretty straightforward perl, and open source. I know I'll be modifying it in the future, and I hope others will find it useful enough to do the same.

      -- TTK

  135. Tahya al-moqawama al-Iraqiyah! by Anonymous Coward · · Score: 0

    Tahya al-moqawama al-Iraqiyah!

  136. The Ultimate Storage Solution? I think not... by ning · · Score: 2, Interesting

    10^15 bytes? Each human on Earth has 3 billion (3x10^9) base pairs of DNA. Assuming 2 bits to encode a base pair of DNA, that means a PetaBox(tm) can only store the DNA of 1.3 million people. So you'd need getting on for 5000 of these (assuming no compression) to store the entire population.

  137. Just nerds seem to reply... by Chreo · · Score: 2, Funny

    ... but honestly, that's a lotta pr0n!

    --

    Life is what happened when Good Intentions met Harsh Reality (the brother of the more infamous Chaos).
  138. Re:wrong by grahamlee · · Score: 2, Informative
    bytes are not SI units, those would be octets

    Actually the SI defines the prefixes irrelevant of units used. Think of the mil ('milli-inch'); how many do you think there are in the inch? If I had a thousand cats I could refer to the set as one kilocat, and hence if I had 1024 cats I could refer to it as a kibicat, Tweety-pie style; note that a cat is not an SI metrological term. Try playing around with the units(1) command sometime; to get a feel for these SI prefixes.

  139. This is a new use... by thrill12 · · Score: 1

    ...for the term "low power" that I certainly never encountered before... :
    "Low power-- 6kWatts per rack, and 60kWatts for the whole system"

    --
    Slashdot: stuff for news, nerds that matter, matter for news, stuff that nerd
  140. a lot to lose by bobblebob · · Score: 0

    That's a lot of storage in one place - a lot of data to lose - what if there's a fire or flood? How do you back up a petaflop of data?

  141. Experiment by Paulrothrock · · Score: 1
    I wonder how fast this would fill a petabyte:
    cat file1.text file2.text > file2.text
    --
    I'm in the hole of the broadband donut.
    1. Re:Experiment by Walkiry · · Score: 1

      Hmmm

      cat: file2.text: input file is output file

      --
      ---- Take the Space Quiz!
    2. Re:Experiment by Anonymous Coward · · Score: 0

      You could, on the other hand:

      while true; do cat file1.txt >> file2.txt; cat file2.txt >> file1.txt; done

  142. Re:wrong by stanmann · · Score: 1

    Ditto for cleave.

    --
    Food not Bombs is a nice platitude but it breaks down when you notice that the Bombees are usually well fed
  143. Imagine a Beowulf Cluster.... by Anonymous Coward · · Score: 0

    Where is the Obligatory Beowulf Cluster joke?

    I can't believe I'm the 563rd AC post, and got to deliver it first. Maybe they all got modded down to -1.

  144. I have to ask... by sirgoran · · Score: 1

    Since many people have already posted or wanted to post the usual "imagine a beowolf cluster of these", I have to wonder if that might be redundant. Wouldn't one of these BE a beowolf cluster?

    If so, then we can all imagine no more!

    -Goran

    --
    Carpe Scrotum - The only way to deal with your competition.
  145. Modified Joke by ArchAngel21x · · Score: 1

    Imagine Groucho from the Marx Brothers standing between this thing and a hot woman.

    "Why this the the most magnificent rack I've..."

    *Looks at girl"

    "No this is the most mag...."

    *Looks back at the Peta Box*

    "No this is the most magnificent rack I've ever seen."

  146. Re:wrong by HTH+NE1 · · Score: 1

    PiB, GiB, MiB (and also KiB) are all 1000. PB, GB, MB, and KB are 1024 due to BINARY!

    Thank you, but you got it perfectly backwards.

    The "i" comes from the second letter of "binary", as in peta-binary-byte, giga-binary-byte, mega-binary-byte, and kilo-binary-byte, also abbreviated as pebibyte, gibibyte, mebibyte, and kibibyte.

    Read the Wiki and the references there.

    I personally recommend qualifying metric measure with the word "metric", i.e. "3.5 metric GB", or "3.5 GB metric" as well as using the more specific binary names and abbreviations as appropriate to eliminate confusion.

    --
    Oh, say does that Star-Spangled Banner entwine / The myrtle of Venus with Bacchus's vine?
  147. Good thing marketing didn't call the it a PetaFile by Anonymous Coward · · Score: 0

    Good thing marketing didn't call the it a PetaFile

  148. Re:So the question is... by symbolic · · Score: 1

    ...not "Is it ready?" It's, "ready for what?"

  149. Re:wrong by adoarns · · Score: 1

    Giga- from Greek, gigas (damn, no greek entities!), which in modern Greek would come out more Klingonish, but certainly not like the letter J, even if you went back to Attic Greek.

    --
    Tenemus pyrobolos atqui jacimus cognitiones.
  150. Mr. President?! by Anonymous Coward · · Score: 0

    Is that you?!

  151. Re:wrong by Elwood+P+Dowd · · Score: 1

    Doesn't matter anyway. Language changes for no good reason. That's ok. The way people say it is the way it's to be said.

    --

    There are no trails. There are no trees out here.
  152. Re:wrong by Marxist+Hacker+42 · · Score: 0

    You're right, I'm wrong, I'm posting this so that this post doesn't get any more replies. I'm exactly backwards- and was unaware of the 1999 IEC standardization.

    I'm still disappointed that they're not teaching young geeks to count in binary in school anymore. Perhaps we need to start teaching graphics entirely in TI99/4A basic (as that's where I got my first exposure to binary- converting, by hand, 8x8 bitmaps into Hexadecimal for redefining the font).

    --
    SJW: a person who perceives an injustice, and while correcting it, commits a greater injustice.
  153. Teaching binary by HTH+NE1 · · Score: 1

    I'm still disappointed that they're not teaching young geeks to count in binary in school anymore. Perhaps we need to start teaching graphics entirely in TI99/4A basic (as that's where I got my first exposure to binary- converting, by hand, 8x8 bitmaps into Hexadecimal for redefining the font).

    Nah, teach shape tables for Apple II series computers. Have the students work out how to draw an image using vector plotting moves, convert them into 3-bit binary codes (occasional no-plot moves to 2-bit), put the sequences in a table, convert them to hexadecimal, then back into decimal to be put in a DATA statement to be READ and POKEd into memory.

    That will eat up a lot of class time and keep the kids out of mischief for awhile. Especially the debugging of the resulting shape table to find whether they made an error in their vector moves, putting them in the table in the wrong order, used the wrong binary code for a vector direction, a bad binary->hex->decimal conversion, or just a typo in their DATA statements.

    Teaching them how to use the *monitor to do the hex-decimal conversion will be later, and even later how to put them into memory directly and BSAVE them to disk to be BLOADed instead of POKEd.

    And if there's enough time, maybe then get into raster graphics using & commands to interface with machine code using lookup tables and indirect addressing modes to efficiently locate a location on the 8-field interlaced graphics screen to store bytes. (Whether there is enough time depends on whether or not you started the semester on how to load and save programs to cassette tapes.)

    Oh, the memories.... what was this thread about again? Oh, yeah. A pity that ProDOS had a partition size limit of 32 MiB, and that there's also a limit to the number of partitions you can have. That petabyte rack would have a lot of unusable space hooked up to an Apple II (apart from a IIgs which could use HFS). Though maybe... you use a dynamically rewritten partition table so you could address the full capacity in 32 MiB chunks! Though drive 0 in the array would wear out a lot faster than any other.

    (Don't sweat the multiple replies, and don't take them personally. Hopefully each one is more generally informative than the last.)

    --
    Oh, say does that Star-Spangled Banner entwine / The myrtle of Venus with Bacchus's vine?
  154. Spicier by Doc+Ruby · · Score: 1

    It probably takes more than just basepairs to make a Spice Girl (even neglecting the simulation of a Thatcherite childhood). There's also the newly-investigated "methylation" info layer, which modulates the genome's potential codes. And if we're just finding out about methylation, what else lies within?

    --

    --
    make install -not war

    1. Re:Spicier by glwtta · · Score: 1

      Oh sure, but the OP specifically said "DNA sequence."

      --
      sic transit gloria mundi
    2. Re:Spicier by Doc+Ruby · · Score: 1

      From theoriginal post: "Store the DNA sequences for the spice girls on it, such that years from now our decendants may know the joys of Scary Spice.". The response estimated the storage required for only the basepairs at 3.5GB-720KB-150KB. I pointed out that the DNA sequence includes info in more than jsut the basepair sequence, such as the variable methylation state of the DNA backbone. So if you want your Spicegirl to be "Scary", and not just scary, you'll need to unzip more than the basepairs.

      --

      --
      make install -not war

    3. Re:Spicier by glwtta · · Score: 1
      I pointed out that the DNA sequence includes info in more than jsut the basepair sequence

      Seems we are really interested in the semantics here - by definition, the "DNA sequence" is just the "basepair sequence" - an ordered succession of basepairs. It's quite obvious that a lot more information that that would be needed to "reconstruct" Scary Spice, but the OP did in fact just mean the DNA sequence. If you are saying that the DNA chromosome holds more information than just the basepair sequence - fine, but the sequence is still just the sequence.

      (BTW, I am not suggesting that it's possible to reconstruct Scary just from her DNA information, even if you include methylation state).

      --
      sic transit gloria mundi
    4. Re:Spicier by Doc+Ruby · · Score: 1

      In point of fact, I just misquoted myself - I had posted that more info than coded in the basepairs is coded in the DNA that produced the Spicegirl. I further distinguished the non-genetic ("nurture") influences, like her childhood, from the genetic. So I'm exploring the genetic info that would need to be stored to generate a Spicegirl, in addition to the basepairs. The semantics aren't nearly as interesting as the genetics.

      --

      --
      make install -not war

  155. Was his name John Frink? by MrBlackBand · · Score: 1
    Frink: Well, sure, the Frinkiac-7 looks impressive, don't touch it! But I predict that within 100 years computers will be twice as powerful, 10,000 times larger, and so expensive that only the five richest kings of Europe will own them.

    Apu: Could it be used for dating?

    Frink: Well, theoretically, yes, but! the computer matches would be so perfect as to eliminate the thrill of romantic conquest.

    --
    "It is difficult to get a man to understand something when his salary depends upon his not understanding it."