Too many big data initiatives are science projects that take months of effort, risk failure and require highly trained data scientists with scarce skills. According to a CSC survey, 55 percent of big data projects aren’t completed and many others fall short of their objectives.Read more...
IBM Research Sets New Record for Storing Big Data
LAS VEGAS, Nev., May 20 — IBM researchers announced they have demonstrated a new record of 85.9 billion bits of data per square inch in areal data density on low-cost linear magnetic particulate tape — a significant update to one of the computer industry’s most resilient, reliable and affordable data storage technologies for Big Data.
At this areal density, a standard LTO size cartridge could store up to 154 trillion bytes (154 terabytes) of uncompressed data — a 62 fold improvement over an LTO6 cartridge, the latest industry-standard magnetic tape product. To put this into perspective, 154 terabytes of data is sufficient to store the text from 154 million books, which would fill a book shelf stretching from Las Vegas to Seattle, Washington.
This new record was achieved using a new advanced prototype tape, developed by FUJIFILM Corporation of Japan. This is the third time in less than 10 years that IBM scientists in collaboration with FUJIFILM have achieved such an accomplishment. The news is being unveiled this week at the IBM Edge conference in front of more than 5,500 attendees.
IBM scientists break Big Data into four dimensions: volume, variety, velocity and veracity and by 2020 these so-called Four V’s of Big Data will be responsible for 40 zettabytes (40 trillion gigabytes) of data. Much of this data is archival, such as video archives, back-up files, replicas for disaster recovery, and retention of information required for regulatory compliance. Because tape systems are energy efficient and more cost-effective than hard disks they are the ideal technology to store, protect and access archival Big Data.
For example, the Large Hadron Collider (LHC) is the world’s largest and most powerful particle accelerator. By the end of the LHC first three-year running period, more than 100 petabytes of physics data had been stored in the CERN mass-storage systems. Most of this data is archived on more than 52,000 tape cartridges of different types, providing scientists with permanent access to data, which could someday answer fundamental questions about the universe.
“Big data has met its match with tape, not only does the technology provide high capacity in a small form factor, it is also reliable for several decades, requires zero power when not in use, is secure in that cartridges cannot be erased at the push of the keystroke and available for the cloud — all at a cost of less than 2 cents per gigabyte and at a greatly reduced operating expense versus disk storage,” said Evangelos Eleftheriou, IBM Fellow.
To achieve 85.9 billion bits per square inch, IBM researchers have developed several new critical technologies, including:
- a new enhanced write field head technology that enables the use of much finer barium ferrite (BaFe) particles
- advanced servo control technologies that achieve head positioning with nano-scale fidelity and enable a 27 fold increase in track density compared to the LTO6 format
- innovative signal-processing algorithms for the data channel that enable reliable operation with a ultra narrow 90nm wide giant magnetoresistive (GMR) reader.
Since 2002, IBM has been working closely with FUJIFILM particularly on the optimization of its dual-coat magnetic tape based on BaFe particles. In this time IBM scientists in Zurich have dramatically improved the precision of controlling the position of the read-write heads, leading to a dramatic increase in the number of tracks that can be squeezed onto the half-inch-wide tape. In addition, they have developed new advanced detection methods to improve the accuracy of reading the tiny magnetic bits, thereby achieving an increase in the linear recording density of more than 56 percent while enabling the use of a reader that is only 90nm in width.
IBM scientists envision scaling magnetic tape to even higher areal densities in the future and will continue to explore novel media technologies. Earlier this month at the 2014 Intermag conference IBM scientists in Almaden, California, have shown that there is potential to continue scaling tape areal densities beyond 85.9 billion bits per square inch. The scientists studied the magnetic properties of a small sample of sputtered media using two specialized test apparatuses. This is an important breakthrough under highly controlled laboratory conditions that may point the way to continue scaling magnetic recording by means of sputtered media once the potential of low-cost particulate media has been exhausted, but much more research will be required.
IBM has a long history of innovation in magnetic-tape data storage. Its first commercial tape product, the 726 Magnetic Tape Unit, was announced more than 60 years ago. It used reels of half-inch-wide tape that each had a capacity of about 2 megabytes. The areal density demonstration announced today represents a potential increase in capacity of 77,000,000 times compared with IBM’s first tape drive product. This announcement reaffirms IBM’s continued commitment and leadership in magnetic tape technology. For more on the history of IBM and magnetic tape storage visit http://www.ibm.com/ibm/history/ibm100/us/en/icons/tapestorage/
For more information about IBM Research visit www.research.ibm.com