Those are just two examples of the promises that storage nanotechnologies hold: combining the functions of memory chips and disk drives on a single piece of hardware that is a fraction of the size of devices today.
Nanotechnology, the science of engineering functional systems at the molecular scale, holds the possibility of billions of infinitesimally small machines working together to build products from the ground up using readily available materials.
Systems in development today could do away with internal disk drives all together as well as the computer boot-up process, instantaneously bringing applications up when a PC or laptop is turned on. Other nanotechnology hardware may allow data to be stored for more than 100 years without having to refresh media.
Most production applications for nanotechnology are now used in reading and writing from storage media that are many times superior to today's storage products at a fraction of the cost. But these developments are prompting storage vendors of all sizes to examine not only how they will manufacture products in the future but what their business models may ultimately look like as a result of the disruptive nature of nanotechnology.
Large and small storage vendors are well into developing storage nanotechnology that promises to shrink by tens or hundreds of times the space required to fit today's data.
IBM has three projects focusing on storage nanotechnologies.
Perhaps the most promising of projects at IBM are carbon annotates, which are molecule-size objects composed entirely of carbon in a cylindrical structure, giving them unique properties. According to Tom Theis, IBM's director of physical sciences, "Nanotubes with diameters of only 1.5 to 2 nanometers possess many times the strength of steel and conduct electricity as both a metal and a semiconductor." Because of these properties, Theis says, "I can't imagine a more aggressive transistor technology right now."
Another way in which carbon nanotubes may be used is in the production of a high density, nonvolatile random access memory chip that could replace dynamic RAM, flash memory and even hard drives. Nantero Inc. in Woburn, Mass., has built prototypes of a chip called NRAM (for nanotube-based/nonvolatile RAM) that is faster than DRAM, as portable as flash memory, and able to provide permanent storage because the wafer uses nonvolatile storage as its basis. "This technology could enable instant-on computers that boot and reboot without delays and eliminate the need for internal disk drives on computers," says Greg Schmergel, CEO and co-founder of Nantero.
IBM invented two foundational nanotechnologies: scanning tunneling microscopy (STM) and atomic force microscopy (AFM), both of which modify materials at the atomic and nanometer scale. Capable of imaging individual atoms and positioning them one by one, these technologies lay the groundwork for manipulating data at a molecular level.
Another subatomic project at IBM is a high-density storage system known as "Millipede," which is being carried out in IBM's Zurich laboratories based on its so-called AFM technology.
Millipede uses AFM in its drive heads to read and write to a polymer surface by creating and removing indentations that are only 10 nm in size. Also known as probe-based storage, "Millipede is able to re-use the same area thousands of times", says Karin Vey, the communications manager at IBM's Zurich Research Laboratories.
AFM technology works by using thermo-mechanical writing to the platter's surface by applying a local force through the cantilever/tip to the polymer layer and simultaneously softening the polymer layer by local heating. Once softening has been initiated, the tip is pressed into the polymer, and an indentation is created corresponding to the logical bit "1." The layer without indentation represents the logical bit "0".
To read the written information, the cantilever originally used for writing is given the additional function of a thermal read back sensor by exploiting its temperature-dependent resistance. While Millipede is a research project, IBM claims it is at a "very advanced stage."
Another subatomic storage project IBM is working on is called storage-class memory. IBM is attempting to use this technology to create cheap nonvolatile semiconductor memory for use in devices like cell phones and cameras.
Gian-Luca Bona, IBM's head of science and technology research, said that though storage-class memory is still in the development phase and no products are yet associated with it, "storage-class memory devices could also be used in the creation of microdrives that close the gap between flash drives and hard drives."
Colossal Storage Corp. in Pokomoke City, Md., is developing a rewritable 3-D volume holographic removable disk media. The nanotechnology under development at Colossal is a possible replacement for today's magnetic disk drives and memory chips. Unlike magnetic media, which only stores data on the surface of the disk drive, holographic optical disk drives use two or more laser beams that work with one another to read and write data throughout the disk media.
Michael Thomas, CEO of Colossal, says holographic optical media drives are superior to other storage nanotechnologies because of their 100+TB capacities, near zero read and write response times and 100-plus year lifespan.
Unlike AFM-based storage nanotechnologies, which require two dissimilar materials to come into contact with each other and create friction and shorten a disk media's lifespan, holographic storage has noncontact surfaces, so it has a higher degree of reliability. "Introducing a media like holographic optical disk drives allows users to invest in a disk media once and for all and not force them to continually reinvest in new storage technologies," Thomas says.
The first generation of holographic optical disk media and disk drives is scheduled to hit the market this year when InPhase Technologies Inc. in Longmont, Colo., releases a 300GB holographic disk and drive. About the same size as today's DVDs, they will hold the equivalent of 64 full-length movies. While they initially will be available in a write-once format, a rewritable disk is on InPhase's product road map.
Despite the promises of storage nanotechnology, advances in existing magnetic disk media technologies and difficulties in constructing reliable production facilities are slowing the development of these next-generation technologies.
HAMR time
Seagate Technologies LLC's new HAMR (heat-assisted magnetic recording) technology addresses current concerns about today's perpendicular recording methods for magnetic disk media. According to Mark Kryder, Seagate's chief technology officer and senior vice president of research, current disk media with perpendicular recording could reach its limits in about five years. "HAMR could extend magnetic recording areal density by about a factor of 10 beyond what can be accomplished with perpendicular recording, and has the potential of extending the hard drive technology another six to seven years beyond its five-year limit."
HAMR uses a laser and a magnetic head together to read and write data on new and more stable disk medium such as iron-platinum. The laser heats the disk medium while the magnetic head writes to it, allowing the disk to store more data. After the media cools, the disk and data becomes very stable. Kryder says, "There are a large number of media that can be written by HAMR and iron-platinum materials can theoretically support 50 terabits per square inch."
Difficulties in building production lines that deliver the right chemical balance that can produce carbon nanotubes are another concern. IBM's Thomas Theis finds these particularly vexing. "You have to learn how to purify the chemical mixes because if you do not get the right balance, you end up with a mix of metal and semiconductor carbon nanotubes," says Theis.
However, Colossal Storage's Thomas firmly believes that some of these technologies could be in production as soon as two or three years from now if the right market drivers were in place. But at the current pace, the soonest any of his projects will even reach the laboratory stage is in 2010, with production starting no sooner than 2012.
So with the potential for devices as small as flash drives to hold as much data in 10 years as the world's largest data centers held only 10 years ago, users like Howard Haile, the director of information systems for Riley County, Kan., see tremendous upside but equally great risks. Haile says, "This would be extremely cost-effective for replicating data off-site for disaster recovery. On the other hand, right now, I don't have to worry about anyone walking out of my offices with my corporate database on a flash drive."
SIDEBAR
Carbon nanotubes: a replacement for hard drives and flash memory?
Carbon nanotubes may be used in the production of a high density, nonvolatile random access memory chip that could replace DRAM, flash memory and even hard drives.
SIDEBAR
The Colossal Storage FE Optical
The Colossal Storage FE Optical Drive will offer symmetrical nondestructive read and writes for the retention of data storage for 100 years or more. Thomas says patents on a semiconductor read/write head for ferroelectric optical storage media memories promise to raise data storage densities by a factor of 1,000 or more and will add at least 10,000 times the data storage capacity per peripheral storage footprint.
SIDEBAR
Heat Assisted Magnetic Recording technology
Seagate's HAMR uses a laser and a magnetic head to read and write data on new and more-stable disk medium such as iron-platinum. The laser heats the disk medium, while the magnetic head writes to it, allowing the disk to store more data. After the media cools, the disk and data becomes very stable.The technology can theoretically support 50 terabits per square inch.