A 70 TB tape cartridge: too much, too late?

A 70 TB tape cartridge: too much, too late?

Summary: IBM and Fujifilm have demonstrated a technology that, if productized, could give us a 70 TB tape cartridge. Is tape dead - or merely sleeping?

TOPICS: Storage, Hardware

IBM and Fujifilm have demonstrated a technology that, if productized, could give us a 70 TB LTO tape cartridge. Is tape dead - or merely sleeping?

Vacuum column, 800bpi tape drives Magnetic tape is the oldest digital storage technology still in use. Once mass storage meant tape storage because drums - and later, disks - were tiny and absurdly expensive. Tape predates disk technology by almost a decade, but it's retreated to niche archiving applications.

They demonstrated a density of 29.5 billion bits per square inch on linear tape. Linear tape uses a stationary head to lay data down in long straight tracks. Video tape has a rotating head that lays down short tracks that angle across the tape.

Linear tape is more reliable for data applications due to reduced tape wear and more accurate positioning. Disks are approaching 1 trillion bits per square inch - but tapes have a lot more area.

If economics and manufacturing technology allow this could lead to a single tape cartridge with a 35 TB of uncompressed data capacity. Since tape drives typically compress data at a 2 to 1 ratio this means 70 TB of data in a single LTO (linear tape open) cartridge.

Current LTO tapes, even with compression, file at about 2 TB per cartridge -- the same as high-end disk drives. In nine months those 2 TB disks will cost about the same as single LTO cartridge. Why store data on disk where it is so much faster to access?

Defenders point to tape's energy efficiency -- write once and shelve without consuming more energy for decades -- but people like the convenience of random-access data. If this drive industry woke up and started offering archive quality disks -- Seagate sold an automotive hard drive that carried a 10 year warranty -- much of the remaining tape market would disappear.

Lifespan is another benefit of tape technology. I recently transferred a 20-year-old VHS tape that hadn't been looked at in at least 10 years to my computer. There was some drop out but the picture was very watchable. Try that with a 20 year old disk drive.

Technology The IBM/Fuji film team attributed their success to four technologies.

  • Advanced nano particle technology -- they limited the size of the barium ferrite particles to 1600 nm3 -- approximately 1/3 of current metal particle volume.
  • Advanced nano coating technology -- a smooth and thin magnetic layer with very low variability reduced signal fluctuation significantly, enabling more accurate signal processing.
  • Advanced nano dispersion -- a new material controlled agglomeration enabling more uniform dispersion of the nano particles.
  • Nano perpendicular orientation -- taking advantage of the barium ferrite particles crystal magnetic anisotropy, a perpendicular orientation improved high-frequency characteristics.

Let's see: mass production of tiny uniform nanoscale particles; mass production of an extremely smooth and thin magnetic layer; and careful control of the particle dispersion and orientation. What could be simpler? Hey, make it cheap too.

A read/write head that can use all that would be a plus, too. Is that too much to ask?

The Storage Bits take Regardless of whether you think tape has a long-term future, this is an impressive demonstration. When I introduced DLT at DEC, we were thrilled to get to 2.6 GB on a tape cartridge.

Like most such demonstrations this has both a marketing message and a technical message. The marketing message is "tape has a lot of life left in it" but the hidden technology message is "it will be many years before you see a 70 TB tape cartridge."

If they can get the cartridge to market in the next 5 years, they'll can charge 5x what a disk costs - because the capacity is so much higher than any single disk. If they can't - well, it was a neat tech demo.

Nevertheless, tape remains the most proven archival storage medium for digital data. Tape may yet live to see that 70 TB cartridge delivered.

Comments welcome, of course. I used an audio cassette recorder for mass storage on my first computer. Couldn't afford $800 for a 144 KB floppy disk. I now have 11 disks - and 2 optical drives - on my Mac Pro. That cassette recorder was my 1st - and last - tape drive.

Topics: Storage, Hardware

Kick off your day with ZDNet's daily email newsletter. It's the freshest tech news and opinion, served hot. Get it.


Log in or register to join the discussion
  • Sounds good.

    Tape dead, you must be kidding, right?
    If you're not using tape you don't really understand backup or point-in-time recovery.

    70TB tape library, heck, I could even back up ALL of our data then. I only use disk-to-disk w/rsync and the secondary array is backed-up to tape. Millions of small files are significant overhead to backup, D2D to tape is a good fit as full backup occurs off line.
  • Oldest digital storage technology?

    I think punched cards and paper tape pre-date magnetic media by many years.

    Presuming of course that "digital storage technology" means mechanisms used to store information in binary form for use in computing devices.
    Fred Fredrickson
    • Still in use.

      Right on both counts - especially if you count the Jacquard loom
      punchcards from the early 1800's. Herman Hollerith put the idea to good
      use in his tabulating equipment for the 1890 census, founding a
      company that later became part of IBM.

      But if anyone is using punchcards today for data storage, I am blissfully
      unaware of it. Likewise paper tape.

      R Harris
      • Or rocks in a desert...

      • Of course punch cards are still around...

        Depending on where you live - you may be voting with punch cards.

        Remember "hanging chad!"
    • not years, centuries...

      The punch card was developed a couple of hundred years ago for use with semi auotmated looms which allowed patterns to be easily and perfectly reporduced.
  • RE: A 70 TB tape cartridge: too much, too late?

    So why did you use the word 'productized'?

    Surely 'produced', 'manufactured', or 'commercially developed' would have been a neater and better choice. It would also have avoided your US-centric tendancy to bastardise the English language.

    And no, Im not British so save the flames.

    • Term of art

      "Productize" is a term of art in product management. It refers to the
      host of activities required to take a technology from concept to a
      useful and successful product.

      Much of the required work has little to do with engineering or
      manufacturing, though they are integral to the process.

      While I dislike the tendency to take perfectly good nouns and turn
      them into verbs, I honestly don't know of a better word to describe
      this process.

      R Harris
      • "Commercialize" works for me

        And it's a real word, too. People have been doing something called "product development" for ages -- somehow they talked about it without inventing anything as godawful as "productize". Yecch.
      • Usage

        I believe this would be the same as using 'weaponize' as in "weaponized anthrax."
    • all language is evolutionary

      otherwise "english" wouldn't exist, its a healthy
      process, all words have their origins as being
      "made up" at some point.
  • What about transfer speeds and contention?

    I consult at an organization that relies on tape for disaster recovery. They recently went to 1TB and the capacity managers decided to reduce the number of tape drives. Although you can hold it all on the tape, accessing it becomes slower as there is reduced parallelism.

    Tape is also part of the hierachial file storage system (unused files migrate to tape). Reducing tape drives increases contention as you increase the chance of multiple processes trying to access different files on the same tape volume.

    As spinning disk, SSD and bandwidth prices continue to drop, tape will lose its place in DR and archive. For truly long term archival, optical probably outlasts tape. You may tolerate some dropped frames watching VHS, but a database backup file with a few dropped bits would be totally useless.

    • Good points -


      You are likely to be proved correct. Producing mechanical devices -
      heads and transports - with the required precision is not cheap. And as
      volumes drop, persuading customers to foot the bill becomes more

      Despite much effort, we do not have a good solution to the problem of
      long-term - 100 year - digital data protection.

      R Harris
      • agreed

        "Despite much effort, we do not have a good solution to the problem of long-term - 100 year - digital data protection. "

      • 100 year storage... Thought.

        Very little data is liable to be needed in 100 years. Do we reall y need to keep every scrap of digital ephemera?

        Periodic purging of old records is still the best way to keep a handle on the stuff. Keeping equipment that can read digital media for 100 years will be a hell of a trick. My advice? Purge the stuff you dont need and make a paper copy of what you do need if you really want to keep it that long. (acid free paper of course)
    • dropped bits totally useless ...

      There are differences in the analogy. VHS is analog opposed to digitized. It generally is more reliable to grab digitized information. Secondly bits can be dropped because ECC is used in hardware. That ECC can do amazing things. Maybe you can recall the "hole punch" demo where somebody would punch a hole or shoot a floppy and the software would still recover all of the data. We used to use a magnetized screw driver and touch the surface of a disk and still recover all the data.

      Also that VHS was probably played many times on a piece of equipment that had played many other VHS and typically these tapes have accumulated tons of dust. This means more wear both on the tape and the equipment. Unless data archival media is being treated with typical lack of care normal in most homes this shouldn't be as much a factor in most data centers. Generally after 20 years the data isn't useful anyway.
  • I've heard of tape still being used.

    I've heard of tape still being used. Some older instructors at colleges insist it's still being used for archival and backup and storing large amounts of data.

    But that's something that I haven't seen at any of the places I've worked, so I wonder about it.
    • Example: the Shoah Foundation uses tape

      to store something on the order of 100,000 hours of video of Holocaust
      survivors testimony. It is several petabytes of data. Every few years they
      transfer the data from old tapes to new.

      But as your comment reflects, the quantity of data required to make tape
      cost-effective keeps growing. It is a shrinking pool, but tape isn't dead

      R Harris
      • What the USC Shoah foundation uses.

        the tapes you refer to are the actual raw
        video taken in the field. These are not the
        same as the data tapes referred to in the
        article. We have about 235,000 individual
        master tapes.

        We are in the process of converting those to
        loss less MPEG2000 files that are stored
        online. Each tape takes roughly 1 GB to store.
        We are no longer copying from tape to tape to
        preserve the video. Most of the copying of
        the tapes in the past were for the
        purpose of combining the masters (the original
        tapes) onto a sub-master tape (usually 4 per)
        that was in turn used to create copies on VHS
        tapes for the catalogers, researchers,
        survivors and the family of survivors.

        We are now using MORE tape as we need it to
        store the files on the back end. In addition
        to the MPEG2000 tapes, we also store
        derivative copies (FLASH, Quick Time, WMV,
        etc.) And likely will be storing new formats
        as they are developed. We are creating the
        files to allow for better access and
        duplication. I just checked with the network
        guy and he tells me we have about 3800 1TB
        tapes in use of which we are using 75%. I
        believe we have converted roughly 25% of our
        tapes to digital files, so this number will
        surely grow over time.

        I'm not writing in any official capacity, just
        relaying information to the best of my

        • I wrote about the Shoah Foundation's archive in

          a ZDnet post called "Long term personal data storage"
          (http://blogs.zdnet.com/storage/?p=376) a year ago after
          interviewing the CTO, Sam Gustman.

          Some key points from that post:
          "They have over 100,000 hours of video on 235,000 tapes. Taping
          started some 20 years ago, so they?ve been dealing with the media
          and format issues for years.

          USCSF is transferring all their original tapes - many in now-obsolete
          formats like BetaSP and VHS - to the 75Mbit motionJPEG2000 format.
          MJPEG2000 is the format chosen by the Library of Congress and is the
          basis for format used in the Digital Cinema Initiative. . . .

          The complete archive requires 8,000 Terabytes of capacity on 2 high-
          end Sun StorageTek tape silos - each costing about million bucks.
          Every 3 years they copy everything to new tapes to ensure

          R Harris