IBM and Sony Cram Up To 330 Terabytes Into Tiny Tape Cartridge (arstechnica.co.uk)
IBM and Sony have developed a new magnetic tape system capable of storing 201 gigabits of data per square inch, for a max theoretical capacity of 330 terabytes in a single palm-sized cartridge. From a report: To achieve such a dramatic increase in areal density, Sony and IBM tackled different parts of the problem: Sony developed a new type of tape that has a higher density of magnetic recording sites, and IBM Research worked on new heads and signal processing tech to actually read and extract data from those nanometre-long patches of magnetism. Sony's new tape is underpinned by two novel technologies: an improved built-in lubricant layer, which keeps it running smoothly through the machine, and a new type of magnetic layer. Usually, a tape's magnetic layer is applied in liquid form, kind of like paint -- which is one of the reasons that magnetic tape is so cheap and easy to produce in huge quantities. In this case, Sony has instead used sputter deposition, a mature technique that has been used by the semiconductor and hard drive industries for decades to lay down thin films.
Putting a few of their numbers together seems to imply a read bandwidth of 24 Tbps (330 terabytes = 330*8 terabits/tape, tape length = 1098 meters => 2.4 terabits/meter, tape speed = 10 m/s => 24 Tb/s). Which to me is way more interesting than the total storage size. Of course I have no idea what you could actually feed that into, it far exceeds CPU memory bandwidth, let alone SCSI or PCIE. Do these things not actually run at full speed for more than a fraction of a second?
Time to download more porn. ;)
Ask me how the Heisenberg Principle may or may not have saved my life.
I figure both of these revolutionary technologies will hit the market at the same time.
Can we get a consumer-level tape drive, something that works with USB 3.x, costs around $1000, with media costing $10-25 a cartridge, with an actual archival life, and some basic features like AES-256 encryption and compression in hardware? Having WORM media would be nice as well.
This would solve a ton of storage/backup issues. A lot of people don't have the network connection to make cloud backups (much less complete disaster recoveries) feasible. People may not trust the cloud. Regulations may prohibit use of a remote backup provider. Or, someone just likes having the peace of mind of physical control of their data on media which might last past the end of this decade.
There are tons of high capacity optical offerings available (Sony has some)... but other than specialty markets, they are not worth it. What would be nice would be a tape or even an optical drive for backups.
Optical wouldn't be bad either. 400 CD jukeboxes are a solved problem, so having a BD-like format with 2-3 terabytes per disk would solve a lot of backup problems, especially with WORM capability to fend off ransomware.
Anyone know how fast these things are? How long would it take to write and how long would it take to read 330 terabytes of data on a similar tape drive?
Yay! Finally enough storage to save our brain on tape! It will be more popular than cryo.
I wonder how better the SNR would be.
While the space would be amazing for backups, if a single head can only write to the media within similar speed ranges of modern tape drives then this tech would be severely limited in actual day to day use in a data center. Large, static data. Which don't get me wrong there is plenty of need for backups for - but it couldn't replace your entire data center's backup strategy with a few drives. At least not without very high write speeds.
There's a problem with that. OTOH... Will the tape fit into my Walkman? I need more storage space for my Napster collection.
Back in the '90's and early 2000's I fell for the 'backup tape' meme three times. Every single time it ended the same way: Buy backup tapes. Put important stuff on them. Put them away. Some time later, you need something off the tape. Whoops, tape drive isn't working right anymore, not even with a new blank tape! Get a new tape drive. Whoops! Not compatible with your old tapes. Or, Whoops! It just won't read your old tapes, even though it's supposedly compatible! Throw out ALL THE TAPES because they're now useless junk and START OVER FROM SCRATCH. Repeat this THREE TIMES.
Hundreds of dollars spent, each iteration.
Never again.
Hopefully, for the first time in years, we'll have a backup medium with the capacity to do a true full backup of everything... that's also fast enough to finish the job in less time than it takes to render the backup almost moot, with enough capacity to include plenty of forward error correction.
Of course, if it satisfies the capacity, speed, and reliability criteria, it'll probably cost at least ten times as much as the computer it's being used to back up, and will thus effectively not exist for non-enterprise users.
QIC was dreadful in every meaningful way. In some ways, the false sense of security it gave was almost worse than having no backup at all.
DDS ("dat") was ok, but by the time it finally became semi-affordable, we were back to needing dozens of tapes to back up a single hard drive.
By the time LTO replaced DDS, its capacity was hopelessly outstripped by hard drive capacity (nominal 800GB tapes, vs 4TB+ cheap hard drives).
Woo-hoo!
D'oh!
#DeleteFacebook
I was going to write a post nearly identical to yours. After about 3 generations of backup tape technology and not ONCE getting anything useful off a backup tape, I quit tape forever. Now I just try to keep my stuff under 1 TB (er, 2 TB, er, 3 TB) and image to a backup drive of the same capacity.
While it would certainly be impressive if it delivered those numbers I strongly suspect that the final product will use a narrow set of heads and require multiple passes to fully access the tape. According to Wikipedia LTO 7 for example requires 112 passes to fully read or write a tape. Also, even if full width heads could be economically built it would be troublesome to sink or source data at that rate, and as far as I'm aware tape drives take a relatively long time to seek and restart streaming after a buffer under/over run.
If you burn at the absolute slowest speed your drive supports, use quality media (m-disc, or archival grade optical... although I've just bought cheap stuff and kept it cool and dark and it has lasted...)
The biggest things to ensuring long term stability of optical archival media is keeping it out of the sun, burning it at a slow rate of speed, and keeping it cool and protected. If you do these 4 things it will last you 20+ years (and yes I have CD-Rs that have now lasted since the late 90s! Y'know back before the scratch protect layers!)
Is that ten years worth of 1080p porn in your pocket, or...
John_Chalisque
Before you get excited about this announcement, note that IBM and Sony has a history of announcing tape drive vaporware. Here is from wiki:
"In 2011, Fujifilm and IBM announced that they had been able to record 29.5 billion bits per square inch with magnetic tape media developed using the BaFe particles and nanotechnologies, allowing drives with true (uncompressed) tape capacity of 35 TB.[18][19] The technology was not expected to be commercially available for at least a decade.
In 2014, Sony and IBM announced that they had been able to record 148 gigabits per square inch with magnetic tape media developed using a new vacuum thin-film forming technology able to form extremely fine crystal particles, allowing true tape capacity of 185 TB.[20][21]"
Even their 2011 announcement has not been brought to market yet.
One step closer to the 300 petabyte Write Only Memory.
That leave a problem unsolved with backups: the time spent to transfer data from one random server to the tape. Even on high speed LAN with top of performance tape, it can take some time.
..it's dead. Tape is dead, Jim. It's a rare backup problem that can't be better solved by remote duplication.
They have worked out how to store 50x as much data on a tape that costs 100x as much to make.