Although the offices of IBM and Hewlett-Packard are nearby, Longmont, CO, is decidedly not Silicon Valley chic. But in this Denver suburb, a radical experiment in data storage is under way. At the headquarters of InPhase Technologies, where the conference rooms are named after ski resorts, chief executive Nelson Diaz holds up a clear plastic disc, about the size of a DVD but thicker, and pops it into a disc drive. A laptop connected to the drive downloads streaming video of an old episode of Seinfeld as the drive writes it to the disc.
But this is no ordinary recording process. The disc has more than 60 times the storage capacity of a standard DVD, while the drive writes about 10 times faster than a conventional DVD burner. That means the disc can store up to 128 hours of video content – almost twice enough for the full nine seasons of Seinfeld – and records it all in less than three hours.
It’s likely to be one of the first commercial systems to use “holographic storage,” in which bits are encoded in a light-sensitive material as the three-dimensional interference pattern of lasers. Unlike CDs and DVDs, which store data bit by bit on their surfaces, holographic discs store data a page at a time in three dimensions, enabling huge leaps in capacity and access speed. And InPhase, a 70-person startup spun out of Lucent Technologies’ Bell Labs in Murray Hill, NJ, is leading a handful of companies racing to commercialize this optical storage breakthrough.
Three-dimensional memory could dramatically change how we use microelectronics. Many of the remarkable advances in consumer electronics over the last few years – and much of the economic health of the industry – are directly traceable to the explosion in storage capacity. Web e-mail services routinely offer each of their customers a gigabyte of memory for free. Apple’s newest iPod is only possible because of small, cheap hard drives that can hold a staggering 60 gigabytes of data – a storage capacity that just five years ago would have been a lot for a desktop PC.
Likewise, cell phones now come with flash memory chips easily able to store address books, calendars, photos, and the like. Meanwhile, CDs and DVDs have already transformed how people listen to music and watch movies. But each of these storage technologies has drawbacks. The density of magnetic materials in hard drives is fast approaching a fundamental physical limit. Flash memory is slow, and a DVD is barely large enough to hold a full-length movie.
Storing data in three dimensions would overcome many of these limitations. Indeed, the theoretical promise of holographic storage has been talked about for 40 years. But advances in smaller and cheaper lasers, digital cameras, projector technologies, and optical recording materials have finally pushed the technology to the verge of the market. And the ability to cram exponentially more bits into infinitesimal spaces could open up a whole new realm of applications.
By storing and reading out millions of bits at a time, a holographic disc could hold a whole library of films. Movies, video games, and location-based services like interactive maps could be put on postage-stamp-size chips and carried around on cell phones. A person’s entire medical history, including diagnostic images like x-rays, could fit on an ID card and be quickly transmitted to or retrieved from a database.
Eventually, if the hardware becomes affordable for consumers, holographic storage could supplant DVDs and become the dominant medium for games and movies. Portable movie players and phones that download multimedia from the Web would take off. Holographic storage could even compete with the magnetic hard drive as the computer’s fundamental storage unit. And on a larger scale, corporate and government data centers could replace their huge, raucous storerooms of server racks and magnetic-tape reels with the quiet hum of holographic disc drives.
InPhase’s competitive edge lies in its partnerships with Hitachi Maxell, a leading producer of computer tapes and CD-ROMs, and – as of this May – Bayer MaterialScience, one of the world’s largest makers of plastics used in optical discs. These large corporations see holographic techniques as the next step in the evolution of storage. “Our collaboration with InPhase gives us a tremendous opportunity,” says Hermann Bach, head of technologies for the Americas at Bayer MaterialScience.
But if and when holographic storage will come to dominate the market is still an open question. InPhase’s initial product launch is slated for late 2006, but industry experts, while optimistic, are also cautious. “They have made numerous contributions on the hardware side, in media and materials, and in error correction,” says Hans Coufal, manager of science and technology strategy at IBM’s Almaden Research Center in San Jose, CA, and an expert on holographic storage. “It’s very impressive but still some ways away from a viable product. Not a long ways, but some ways.”
The idea of holographic storage dates back to the work of Polaroid researcher Pieter J. van Heerden in the early 1960s (and, some contend, to Nobel laureate Dennis Gabor’s original theory of holography in 1948). But the technology had never been practical, requiring exceedingly expensive materials and bulky laser setups – unlike the streamlined system from InPhase. Even Bill Wilson, InPhase’s chief scientist, was originally skeptical. In 1987, as a fresh PhD in physical chemistry from Stanford University, Wilson joined Bell Labs, turning down a job at IBM, where he would have started working on holographic storage. “I thought the field would be a complete waste of time,” he admits.
The turnaround began in the early 1990s, when IBM and other big players started to worry about the limitations of magnetic storage. As storage capacity increases, the magnetic grains that store data on a hard drive get packed closer together. Eventually, each grain’s magnetic field will begin to interfere with those of its neighbors, hindering their ability to reliably hold data. Engineers have thought of clever ways to defer this problem, but ultimately, grains in magnetic materials will be too dense to work properly.
Wilson recalls jumping into a friendly argument in the Bell Labs lunchroom about what new technology could eventually take the place of magnetic media – and the relative merits of holographic storage. At the time, the technique was undergoing something of a revival, being investigated by research groups at IBM, Polaroid, Caltech, and Stanford. Wilson and Kevin Curtis, an electrical engineer from Caltech who had recently joined Bell Labs, argued that holographic storage might actually become viable with suitably small and cheap optical components. In discussing the technical issues with their colleagues, they realized the key to making it viable was the material that stored the data.
In holographic storage, a “data beam” holding information is crossed with a “reference beam” to produce an interference pattern that’s recorded in a light-sensitive material. To retrieve data from a particular spot, a reference beam is shone onto it, and the combination of the reference beam and the patterned material reconstructs the original data beam, which is read by a digital-camera detector that translates the beam into a series of electrical signals. The recording material is typically either an inorganic crystal or a polymer. Polymers are more sensitive and require less powerful lasers, but they have their own flaws. For instance, when you hit a photosensitive polymer with a laser, it tends to deform, which messes up the data.
In 1994, a materials team at Bell Labs led by chemist Lisa Dhar worked with Wilson and Curtis to produce a “two-chemistry” photosensitive polymer. The researchers mixed one scaffoldlike polymer, which stayed rigid and preserved its structure, with another polymer that reacted to light and stored data. Decoupling the recording material’s optical and structural properties let the researchers fine-tune each independently, arriving at a combination of sensitivity and stability that had eluded previous efforts.
Over the next four years, the Bell Labs team got its holographic material to work in conjunction with the latest miniaturized lasers, cameras, and optical components to read and write data. This also required advances in software to correct for errors in storing and retrieving digital bits. In 1998, as a proof of concept, they built a prototype holographic recorder and recorded MP3 digital audio in real time. It was a bulky contraption and not particularly efficient. But at that point, says Wilson, “we realized we could build the darn thing.”
So in mid-2000, the researchers contacted Nelson Diaz about starting up a company. Diaz had made his name in the storage industry, working as an engineer for nearly 20 years at Digital Equipment Corporation and most recently as a general manager at StorageTek in Louisville, CO, a leading maker of disk and tape drives. When first told of the researchers’ focus on holographic storage, he was skeptical: he had heard the hype for years. But the closer he looked at the Bell Labs design, the more he believed. Five months later, he signed on as chief executive of InPhase.
The first order of business, says Diaz, was getting rights to the underlying intellectual property. InPhase negotiated a deal with Bell Labs that gave it ownership of the core patents for the holographic storage system. Then, of course, the company needed funding. In late 2000, before the tech bubble collapsed, InPhase raised $15 million in three weeks “without a business plan,” says Diaz. (Storage giant Imation was a first-round investor.) So in December 2000, six researchers from Bell Labs, including Wilson, Curtis, and Dhar, moved out of the suburbs of New Jersey and joined their new CEO in Colorado.
Four and a half years later, the company is still working to develop a holographic storage product, explains Demetrios Lignos, InPhase’s vice president of engineering. Lignos is another veteran of the storage industry, a bottom-line guy, not one to be impressed by fancy science or research demos. Product development, he says, takes time; in this case, the challenge was shrinking the optical components down while maintaining the insane levels of precision needed to make holographic storage reliable. Now his team of 60 engineers is gearing up for a pilot launch in September 2006 and, if it goes well, a full release to follow. The initial product: a holographic disc drive that reads and writes 300-gigabyte discs.
But don’t throw out your hard drive just yet. The cost of InPhase’s holographic equipment will be beyond the means of consumers and most digital-content distributors for some time.
Sitting in front of six holographic disc drive prototypes, Lignos explains what makes them tick. Inside each breadbox-size drive is an elaborate system of mirrors, lenses, and liquid-crystal displays that manipulates the beam from a single laser. The disc, 130 millimeters in diameter and 3.5 millimeters thick (as compared to 120 millimeters and 1.5 millimeters, respectively, for a DVD), doesn’t spin continuously like a DVD but is mounted on a stage that positions it so that the right portion is exposed to the laser beams at the right time. The laser and camera detector are fixed, but the mirrors and lenses move to produce different beam angles. And that’s the real trick: unlike a CD or DVD, the disc can store hundreds of pages of data in a single, small area, each one inscribed by the reference beam at a slightly different angle.
The technology is here. The question now is the size of the market. “Will it actually get into the hands of many users? We haven’t proven that yet,” acknowledges Lignos. For InPhase, the first applications will lie in high-end archiving for data centers, financial institutions, and medical centers. In those markets, holographic storage will compete with magnetic tape, which also has a high storage capacity but is harder to access. It’s also less durable, lasting less than 10 years, while holographic discs should last 50 years or more.
InPhase also plans to go after high-definition digital video broadcasting and movie distribution for digital theaters: companies such as the Turner Broadcasting System want to archive videos; and one can imagine the next George Lucas extravaganza being delivered to digital cinemas on one disc instead of a stack of 100.
By 2007, InPhase plans to release a consumer electronics product, a chip that could hold up to five gigabytes – enough to store a movie or video game. The chip could compete with flash memory and give handheld devices the ability to quickly download and play back high-resolution content on the fly.
InPhase is focusing on video games, where there are fewer global standards than in movie distribution – making it easier for a small company to break in with new technology. And holographic discs have an advantage for content distributors: they are difficult to pirate. Creating a copy requires the same expensive equipment necessary to make the original.
Five to ten years out, holographic storage could become a mainstream consumer technology – or a colossal flop. The still unanswered questions involve the long-term reliability of the components and, of course, cost. The technology must be dependable enough to convince customers to trust it with their most important data yet cheap enough to become ubiquitous.
InPhase will compete with a smattering of other holographic-storage companies. Tokyo-based Optware is targeting consumer video applications with a simpler technology more similar to traditional DVDs. And Aprilis in Maynard, MA, a Polaroid spinoff, is going after some of the same markets that InPhase targets but is also branching out into biometrics applications like fingerprint matching.
“I expect them to coexist for a while, until the better one wins,” says IBM’s Coufal, an industry veteran who adds that the different companies’ approaches are all appealing. “Everybody would love it to succeed….Who will win, I don’t know.”
But whoever wins, holographic storage could change the rules for information technology by opening up the possibilities of working in three dimensions. Until now, storage – indeed, all of microelectronics – has played out mostly on the surfaces of materials. The benefits of exploiting the third dimension could go beyond storage to include more efficient ways to search ultradense databases, like those that store satellite images for mapping and surveillance; new kinds of displays; and even ultrafast processors whose logic circuits are carved into holographic materials.
“It will take time and some deep pockets,” says InPhase’s Lignos, “but we finally have the ability to take this to market.”
Going bald? Lab-grown hair cells could be on the way
These biotech companies are reprogramming cells to treat baldness, but it’s still early days.
Meet Altos Labs, Silicon Valley’s latest wild bet on living forever
Funders of a deep-pocketed new "rejuvenation" startup are said to include Jeff Bezos and Yuri Milner.
Meta’s new learning algorithm can teach AI to multi-task
The single technique for teaching neural networks multiple skills is a step towards general-purpose AI.
A horrifying new AI app swaps women into porn videos with a click
Deepfake researchers have long feared the day this would arrive.
Get the latest updates from
MIT Technology Review
Discover special offers, top stories, upcoming events, and more.