IBM and Fujifilm Set Tape Storage Density Record
Companies pack unprecedented 123 billion bits on square inch of magnetic tape
April 10, 2015
IBM and Fujifilm announced that their joint research teams have in a laboratory put 123 billion bits of uncompressed data per square inch on magnetic tape, which is the equivalent of 220 TB on a single tape cartridge.
Once thought to have been almost an obsolete technology, Mark Lantz, manager of exploratory tape at IBM Research says, IBM has been investing in magnetic tape research because of the advent of Big Data. Many organizations today are starting to save massive amounts of data to drive Big Data applications. The most economical way to archive that data is still magnetic tape.
Lantz adds that magnetic tape drive systems are also getting faster. Lantz says tape systems can not only transfer 360MB per second, tape storage systems now have much more granular control over where any piece of data is stored on magnetic tape.
As a result, Lantz says tape storage has a role to play both inside traditional data centers and in the cloud.
“We see a huge opportunity to use tape in the cloud,” says Lantz. “We’re now storing data at the nano level on tape.”
Lantz says IBM is also exploring adding support for advanced tape technologies to object storage systems such as OpenStack Swift, which it demonstrated this week at the National Association Broadcasters 2015 conference.
To achieve this level of density IBM and Fujifilm created a new set of servo control technologies that include a high-bandwidth head actuator, a servo pattern, a servo channel, and a set of tape speed optimized H-infinity track follow controllers that together enable head positioning with an accuracy of better than 6 nanometers. This enables a track density of 181,300 tracks per inch, which is a 39-fold increase over existing LTO6 tape technology.
The two companies also developed an enhanced write field head technology that enables the use of much finer barium ferrite (BaFe) particles and new signal-processing algorithms for the data channel that are based on noise-predictive detection principles that enable usage of 90nm wide giant magnetoresistive (GMR) reader.
Lantz says IT organizations in the years ahead should expect to see the amount of data that can be stored on magnetic storage to continue to grow. The only really limiting factor, says Lantz, is the need to maintain backwards compatibility with existing tape technologies.
Citing data compiled by the market research firm Coughlin Associates, IBM says there is already over 500 exabytes of data stored on tape systems today. The irony of Big Data is that many data scientists now want to be able to access that archival data, sometimes known as “dark data” to establish historical trend lines within new Big Data analytics applications. The end result is a lot more loading of tape drives into storage systems that hopefully have actually stood what is amount to a real test of time.
About the Author
You May Also Like