2.8 PETABYTES!?!?!

At InfoDocket there is a story describing the project to digitize the Vatican Library.

Get this!

EMC Corporation has today announced that it is providing 2.8 petabytes of storage to help the Vatican Apostolic Library digitize its entire catalogue of historic manuscripts and incunabula (a book or pamphlet printed before 1501). One of the oldest libraries in the world, the Vatican Apostolic Library holds many of the rarest and most valuable documents in existence including the 42 line Latin Bible of Gutenberg, the first book printed with movable type and dating between 1451 and 1455.

Do you remember “Doc’s” reaction to how much electricity was needed to power the DeLorean?  We are all used to hearing “giga-” these days.  But this is “peta-”

Our hard drives are now in gigabytes and terabytes.  I remember when having megabytes was a big deal.

The prefixes indicate multipliers.  kilo-, mega-, giga-, tera-, peta- etc.

Prefix Symbol(s) Power of 10 Power of 2
yocto- y 10-24 *
zepto- z 10-21 *
atto- a 10-18 *
femto- f 10-15 *
pico- p 10-12 *
nano- n 10-9 *
micro- m 10-6 *
milli- m 10-3 *
centi- c 10-2 *
deci- d 10-1 *
(none) 100 20
deka- D 101 *
hecto- h 102 *
kilo- k or K ** 103 210
mega- M 106 220
giga- G 109 230
tera- T 1012 240
peta- P 1015 250
exa- E 1018 * 260
zetta- Z 1021 * 270
yotta- Y 1024 * 280
* Not generally used to express data speed
** k = 103 and K = 210 
FacebookEmailPinterestGoogle GmailShare/Bookmark

About Fr. John Zuhlsdorf

Fr. Z is the guy who runs this blog. o{]:¬)
This entry was posted in Just Too Cool and tagged , . Bookmark the permalink.

31 Responses to 2.8 PETABYTES!?!?!

  1. VexillaRegis says:

    Impressive!
    Welcome to GMT-1, Fr. Z!

  2. robtbrown says:

    All a consequence of Moore’s Law, which says that the number of transistors on a chip doubles every 1.5 to 2 years.

  3. JonPatrick says:

    The first computer I ever used had 16 kilobits of memory (bytes hadn’t been invented yet). Hard to believe, today even pocket calculators have more than that.

  4. Kerry says:

    All the prefixes sound like tried and rejected Marx brother’s names. (And I see the nickname of a president in the list too). Heh.

  5. robtbrown says:

    We distinguished bits from bytes. A bit is one switch (transistor). Bits were combined to form bytes–one byte being one character, which then could vary from 3 to 8 bits, depending on the system. They were combined form a computer word, which in IBM was 32 bits (another system I worked in had 40 bits/word).

  6. robtbrown says:

    JonPatrick says:
    The first computer I ever used had 16 kilobits of memory (bytes hadn’t been invented yet). Hard to believe, today even pocket calculators have more than that.

    Was it a Sinclair? My first computer was an Amstrad PPC 512, bought in 1989. For 3 years every paper I wrote in Rome was written on it. I had to load DOS every time I turned it on.

    http://en.wikipedia.org/wiki/PPC_512

  7. Phil_NL says:

    Hard to imagine now that towards the end of the 80s and early 90s, one was more likely to see the prefix ‘mega-’ in reference to megatons (TNT equivalent in the yield of nukes) than megabytes.

    Not to mention the fact that ‘exa-’ seemed more than enough to show off your knowledge of multipliers. At current pace, we’ll see that appear in a year or four – now we have to wonder what comes after ‘yotta-’. Anyone with here sufficient greek to count to 9 and 10? ;)

  8. The Masked Chicken says:

    As I tell my students, Yocto and Zepto are the two missing Marx brothers. They never seem to get the joke.

    Now, there has been a move to assign a name to 10^27. The exo, zetta, and yatto, are, x, y, z all dressed up. The Zuhlsdorf should a unit for something, shouldn’t it :)

    Want to remember these? Here’s a mnemonic for 10^15 to 10^-15:

    Please tell grand ma King Henry died unexpectedly drinking chocolate milk. Royal nurse Peterson fainted.

    “Unexpectedly,” is the base unit without a prefix.

    How many Library of Congresses is this? The fictitious unit, LOC, was whimsically defined in 1997 and is 3 petabytes. So the Vatican Library is .933LOC or 93.3% the size of the U. S. Library of Congress. Not that impressive when you consider that each DAY, 70LOC are generated on the Internet!

    The Chicken

  9. tjmurphy says:

    They should make a name change… For the vatican library instead of petabytes they should call them PETRA-bytes :-)

  10. yatzer says:

    Keeping that stuff straight gives me a headache, but I do remember being quite pleased at having a 300 baud modem that operated holding my telephone hand receiver.

  11. JonPatrick says:

    robtbrown, it was a GE 225 a 1960′s era mainframe, at my school (Lehigh University). You submitted programs on punched IBM cards through a slot in the glass wall behind which was the computer, tended by an operator. Your program took several minutes to run as the compiled binary had to be written out to a mag tape, then rewound and loaded. Later on a printout would be placed in a bin, usually with an error. “job card error” or a compiler error. Sigh. Wait in line for the keypunch, retype the card or cards, then resubmit. Rinse, repeat, until done.

    By the way that memory was real “core” memory, made of little ferrite cores strung on fine wires by very patient ladies at the GE factory.

    My first personal computer was the Commodore 64 with a whopping 64K bytes and a cassette tape to store programs. The next generation after the Sinclair I think.

  12. PomeroyonthePalouse says:

    This * Not generally used to express data speed referring to the multipliers less than 1. They obviously haven’t met the company who provides me my internets.

  13. Jeannie_C says:

    Jeannie_C’s other half here.. 35 year IT career, first system I got to play with was an IBM 1401, mainframe with 16K core, then eventually was entrusted with being allowed to sit and operate a 360/40.. I still have a paper tape containing a routine I wrote that “automated” the manual process of de-allocating the memory on the 360 before running another job (no OS auto memory allocation control there!).. After that it was the whole 370 family where I fell in love with VM/370.. after that, storage, disaster recovery, systems management, servers, etc.. Last project before retirement was working on predictive failure modeling using various system logs to try and “predict” and mitigate potential problems/failures. Interesting math on that one.
    BTW, using OC192 (10 Gbps) optical fiber, and assuming NO overhead,it would take over 20000 hours to “download” a petabyte..

  14. APX says:

    I still remember when Napster first came out and I was trying to save mp3s to a 3.5″ floppy and get frustrated that the file size was too big. And then my parents decided to get with the times and bought a CD burner. Wow, we were cruising in the 21st century at a writing speed of 1X and that every attempt to burn a CD was preceded by the prayer that no error should occur in the burning process. An intense 74 minutes. Now my collection of burnt CD’s are obsolete and I’ve moved onto multiple GB memory sticks that used to cost $40+ dollars, now available at Walmart for $5.

    This post just brought backslashes to grade 10 science and the pmeumonic device I had to remember to order of prefixes from terra to pico- The Girl’s Mother Killed Her Dad [ ] Death Can Make Mom Not Pretty. Unfortunately I was not aware there were more suffixes. Now I need a new pneumonic device.

  15. Charles E Flynn says:

    Here are some more details about EMC and the Vatican project:

    EMC Brings the Vatican Apostolic Library into the Digital Age

  16. robtbrown says:

    JonPatrick,

    I started also with card to tape–in 1978. We wrote programs on grid paper, handed them in to key punch, who made two copies of the cards for quality control. Then they were taken out to the system. It would sometimes take a week before we knew whether the compile was successful. There were 3 old Honeywells and 1 new IBM 370. The place believed in dual vendors and tried to replace the old H’s with a new. For 3 months we were told that the new H would be there in one month. Finally, they said forget about the new H–a new IBM was arriving the next week.

  17. robtbrown says:

    I just want to add that there has been no experience in my life like reading a hexadecimal dump.

  18. Mike Morrow says:

    Bumper sticker on IBM 360/370 user’s car: HONK IF YOU LOVE JCL

  19. The Masked Chicken says:

    “I started also with card to tape–in 1978.”

    I started with rocks. Had almost no memory and not even a shift function, but the computer was practically indestructible. So are slide rules, which I think are the height of personal computation. I still use mine. I have a collection going back to a 1947 (?) maple/ivory rule. As I tell my students, the two most important discoveries in physics of the last century – Relativity and Quantum Mechanics were made with thought, pencil and paper, and a slide rule. Heck, we put a man on the moon using slide rule technology. Computers? Ha!

    Still, you haven’t lived until you’ve worked with graphics on a Silicon Graphics workstation, however. They used these to animate Jurassic Park.

    The big news, however, is not the petabytes, but the fact that to scan everything, they are going to have pull the objects from the shelves. Who knows what discoveries that might lead to.

    The Chicken

  20. APX says:

    I still play Space Invaders on my grandma’s Commodore 64. She asked how she could set it up to send emails.

  21. Jeannie_C says:

    Mike Morrow:

    HONK! … IEFBR14 … HONK!

  22. acardnal says:

    The Chicken wrote, “So are slide rules, which I think are the height of personal computation. I still use mine. I have a collection going back to a 1947 (?) maple/ivory rule. As I tell my students, the two most important discoveries in physics of the last century – Relativity and Quantum Mechanics were made with thought, pencil and paper, and a slide rule. Heck, we put a man on the moon using slide rule technology. Computers? Ha!”

    THAT is well worth remembering. I’m pleased you remind your students of that.

  23. Kathy C says:

    That much storage is worth a lot. God bless EMC for such a valuable gift.

  24. That is a lot of storage…in my physics class we don’t get that high on the scale usually.

  25. JonPatrick says:

    RE Slide rules – got through 4 years of Engineering school at Lehigh with one. Computer turnaround took too long (see previous post) for anything but really complex calculations. I still have it somewhere (Faber Castell duplex trig-log).

    By the time I got to grad school after a 4 year tour of the Southwest courtesy of Uncle Sam, programmable calculators were cheap enough even for a grad student on the GI Bill.

    I’m reminded of the scene in “Apollo 13″ where a flight controller is using a slide rule to do calculations when they are dealing with the crisis. Amazing how primitive things were by today’s standards yet how much was achieved.

  26. AnAmericanMother says:

    My husband the Georgia Tech graduate (but a scientist, not an engineer) still has his big Pickett slide rule AND his K&E. With leather holsters. I can multiply on them (he taught me) but that’s about it. I was a history major, for pete’s sake!
    He makes a good point about slide rules that seems sensible to me: because you don’t have decimal points on a slide rule, you have to know the magnitude and be able to estimate about where your answer will wind up. So you have to have a general grasp of the problem.
    Nowadays you just punch in a bunch of numbers and it spits out an answer, so there’s no guarantee that you understand the WHY of what you just did.

  27. Patrick L. says:

    To distinguish between prefixes that use base 10 and those that use base 2, many in the computer world have adopted a relatively new system of prefixes. The new system changes the prefix slightly and appends a lowercase “i” to the symbol.

    For example,

    PB = 1 Petabyte = 10^15 bytes.
    PiB = 1 Pebibyte = 2^50 bytes.

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

    @tjmurphy: +1 on the Petrabyte. Nice. (The comboxes need a “Like” button.)

  28. frjim4321 says:

    Interesting.

  29. JabbaPapa says:

    exabytes are already in use in the descriptions of some extremely heavy *theoretical* database needs — and really, the petabyte will quite possibly become part of some ordinary home PC configurations within decades.

    The largest commercially developed hard drive technology so far would allow 60 terabyte hard drives (NOT on sale yet BTW — and in the nearer future, the hard drives rolling out first will be 6 TB, 8 TB and so on) (so that a home PC owner could theoretically create a storage area in a home PC totalling hundreds of TB in the fairly near future), and the largest storage space created commercially so far (by IBM) reaches 120 petabytes.

    NASA is however building a 1 yottabyte storage space for its own computing needs.

    Last year either Bill Gates or someone else at Microsoft pointed out that the day is now foreseeable that the yottabyte will be too small to describe the heaviest foreseeable single database sizes :-)

  30. Andy Lucy says:

    To all of the sliderule afficianados, at the risk of revealing my inner geek… are any of you guys members of the Oughtred Society? It is a society devoted to collecting and using slipsticks. I still use my old faithful K&E Log-Log Duplex Decitrig every day…

  31. Rich Leonardi says:

    Thanks for spreading the news, Father. It’s my proudest moment as a thirteen-year EMC employee.