Problem solve Get help with specific problems with your technologies, process and projects.

Of pixie dust and paramagnetism

Recent announcements by IBM about the use of "pixie dust" to increase hard disk storage capacity may come as no surprise.

Of pixie dust and paramagnetism
By Jon William Toigo

If you are among the many who have always suspected that "fairy magic" has a lot to do with magnetic data storage, recent announcements by IBM regarding the use of "pixie dust" to increase hard disk storage capacity may come as no surprise.

That's right. The company that first "discovered" how to store data on a rotating aluminum platter -- by coating it with the same quick-oxidizing primer that was used to turn the Golden Gate Bridge such a lovely shade of rust -- has finally come clean about its long-standing relationship with the ethereal world of fairies, gnomes, brownies and other spritely fellows.

Rumor has it that the first disk drive was not actually an IBM innovation at all: kindly elves constructed the RAMAC one night back in 1955 while the vendor's overworked research scientists were asleep in the back room. IBM took the credit anyway and got the patent, alienating the elves in the process.

They subsequently took up residence at Keebler until the recent acquisition of that company by Kellogg. To make a long story short, the elves -- unhappy with the prospect of sharing the limelight with Kellogg's loud-mouthed icon, Tony the Tiger -- decided to return to their quieter digs in San Jose. They brought with them a bag full of ruthenium that could be layered between that plain old platter coating of cobalt-platinum-chromium-boron (CoPtCrB) to create an antiferromagnetically coupled (AFC) media that could push back the effects of superparamagnetism a tad and enable areal densities of up to 100 gigabits per square inch.

In the process, they helped Currie Munce, director of storage systems and technology and head of the IBM Almaden Research Center in San Jose, CA, to save face.

Hard disk drives: Back to the future?
In April 2000, Munce, together with his peers at Hewlett Packard, Quantum and Seagate Technology, stated in a Scientific American article that "hard disk technology" -- meaning platter media, read/write heads, actuator designs, and signal processing algorithms that were either already in production or soon to arrive -- would be capable of supporting areal densities of up to 150 gb/square inch. The claim was important because, based on current rates of disk capacity growth, the industry faced the very real prospect of reaching hard disk density limits of 30 to 40 gb/square inch as early as 2002.

To assuage concerns about impending limits to magnetic disk storage capacity -- especially at a time when pundits were claiming that the amount of data being stored by businesses was growing exponentially -- vendors did something they were otherwise not inclined to do: agree. They agreed that, barring any unforeseen innovation or breakthrough, the hard disk still had enough run way to support IT storage appetites into the middle of the first decade of the 21st Century. That would hopefully provide sufficient time to ramp up unconventional technologies such as perpendicular recording, patterned media, laser- or heat-assisted recording, or really exotic stuff based on holographic or atomic resolution storage.

IBM, AFC and SPE . . .
AFC media is one step toward realizing Currie's claim. It can be viewed as a hedge against the superparamagnetic effect (SPE) -- loosely described as the point at which the magnetic energy holding bits in their recorded states becomes equal to the thermal energy generated in drive operation. When the two energy levels become equal, bits begin to "flip" at random, making stored data unreliable.

In general, the smaller that bits become -- that is, the higher the storage density of disk media -- the less magnetic energy that each bit contains, and the more susceptible the recorded data becomes to SPE-induced bit flipping. AFC media addresses this dilemma by enabling more bits to be reliably written to (and read from) a fixed amount of disk space, but in a more thermally stable way. The technical details are available at IBM Research News at

The bottom line is that -- with the help of some elfin magic -- Mr. Munce and IBM are validated and disk drive technology has obtained some much-needed room for growth.

What will AFC mean to today's IT shop?
There is a lot of good news surrounding AFC. For one thing, AFC raises the bar on areal densities for conventional hard disk drives. That, in turn, enables the perpetuation for a few years at least, of the 120 percent per year improvement in disk capacity, accompanied by a 50 percent per year reduction in price per GB. This yearly improvement in drive capacity coupled with a steady reduction in price is something consumers have come to depend upon as a predictable factor in storage acquisition planning.

The not so good news is that the improvement in capacity does nothing to encourage IT shops to get serious about storage management. The past couple of decades have already demonstrated that, for as long as storage is cheap and plentiful, consumers will have little incentive or interest in managing its contents efficiently. Management generally becomes a hot button only when resources become scarce and prices rise, providing the impetus for administrators to figure out strategies to do more with less.

About the author: Jon William Toigo has authored hundreds of articles on storage and technology and is our resident expert on storage management issues. Toigo is also the author of storage books, including, "The Holy Grail of Data Storage Management."

Additional Resources:
* Share your comments with your peers directly in either our Storage Networking or Storage Management discussion forums. You may even get a response from Jon directly. To post to our Storage Networking forum, go to

* To post to our Storage Management forum, go to

* If you have a specific storage management question for Jon, send it to him in our Ask the Experts feature at

This was last published in June 2001

Dig Deeper on Storage optimization

Start the conversation

Send me notifications when other members comment.

By submitting you agree to receive email from TechTarget and its partners. If you reside outside of the United States, you consent to having your personal data transferred to and processed in the United States. Privacy

Please create a username to comment.