A Phase Change for Memory

Non-volatile memory (NVM), memory that can hold data without the need of a constant power supply, has been critical to the explosion of devices such as cell phones, PDAs, digital cameras, and media players. Its consistently shrinking size and price has married non-volatile memory to the devices it serves; the desire for smaller devices driving the design of smaller chips, and smaller chips driving the design of smaller devices. In this arena there is a clear leader; Flash memory currently rules the roost. Semiconductor International says that though, "several technologies compete for a market segment that grows faster than the entire semiconductor market: the market of non-volatile memories," Flash, "is currently the overwhelmingly dominant NVM technology. Besides its economic advantage, it offers high-density, fast-reading access and is electrically erasable." Flash has been able to retain its lead through constant evolution and refinements, and in becoming the clear leader, Flash has made itself a fast moving, hard to catch target. However, though the public's desire for smaller devices is not expected to diminish, Flash has inherent limitations which may make it impossible to hold onto its position as the leader forever.

Though Flash has been able to continually shrink to fit the needs of the market, it is only possible for it to get so much smaller; ultimately it can only reach 45 nanometer level before its data starts to leak at unacceptable degree. When Flash hits this brick wall, though it may be years off, a successor will be needed. According to Semiconductor International, "The growing importance and need of NVM, as well as the limited performance and anticipated scaling issue of the now standard flash technologies, will continue to feed the development of new memory concepts and materials." This need has not been kept secret, major vendors such as Samsung and Intel are already investigating and prototyping Flash's replacement. Taking the seat, according to Beta News, "has the potential to be extremely profitable; sales of the popular memory solution are now a $18.6 billion business."

Many in the know suggest that the next top dog will be a type of memory known as Phase Change Memory (PCM). This thought is based largely on the fact that PCM has the ability to perform faster than flash and can do so far beyond the size at which Flash stops performing. Also, in contrast to the flash design, in which data cannot be addressed one bit at a time but only in larger blocks of data, PCM will be addressable at the bit level. The New York Times says that, "Such a capability means that the new memories will be more flexible than Flash memory and can be used in a wider variety of applications and computer designs." Electronic Design points out another limitation of Flash, "Flash memory cells degrade and become unreliable after being rewritten about 100,000 times. This is not a problem in many consumer uses, but is another show-stopper for using flash in applications that must be frequently rewritten, such as computer main memories or the buffer memories in networks or storage systems."

With these ideas in mind chipmakers are actively exploring the PCM option. According to TechWorld, "Samsung has already revealed details of its phase-change RAM (PRAM) switching 30 times faster than Flash. Commercial availability is expected in 2008." Intel and ST Microelectronics are collaborating on their own solution, this could be seen as soon as this year. By 2010 Elpida is expected to start commercial production of PRAM. PCM, under development for years, recently leaped into the headlines when a company announced it had developed an alloy that could operate 500 times faster than Flash and at sizes far smaller than Flash could ever reach. Perhaps of equal weight was the name of the company involved, IBM. Along with two computer memory manufacturers, Qimonda and Macronix, IBM, which left the memory race years ago, may now be getting back in; not only to with a material that could unseat Flash, but that some say could give the hard drive market a run for its money.

The IBM team over the course of years of research designed a new semiconductor alloy derived from materials currently used in optical storage devices like CDs and DVDs. Some are suggesting that the discovery may give the companies a huge head start when it comes time to replace Flash. Though Intel and Samsung, both very much in the memory market, have already prototyped Flash replacements, IBM says "the new material has performance advantages over alloys now in use in prototypes made by others in the industry." According to the New York Times, "The advantage of the new material, according to the scientists, is that it can be used to create switches more than 500 times as fast as today’s Flash chips. Moreover, the prototype switch developed by the scientists is just 3 nanometers high by 20 nanometers wide, offering the promise that the technology can be shrunk to smaller dimensions than could be attained by Flash manufacturers." As fast as the chips may be, and despite the changes they could bring to the industry, they won't appear overnight; experts believe it could be at least five years before consumers see the results of the research. Placing a release date so far in the future has some analysts wondering. Just as Flash will hit a wall at 45nm IBM's innovation might hit its wall by placing its time-to-market is so far in the future. Combine that with the fact that Flash is already a technological institution and many will agree that the future of small drives is anything but certain.

Though the Motley Fool says, "The only question remaining is whether the technology is scalable," and suggests that experts say the prospects are good, there are others who have more doubts. PC World says that, "While its prospects appear bright, the technology faces several hurdles." Rob Lineback, a senior market research analyst at IC Insights said in a TechNewsWorld article, "It seems like every time that people think Flash is going to run out of steam, companies find a way to extend Flash . . . This is a promising development, but it will be a couple of generations -- three to six years -- before we see if it's going to be any major force in the marketplace."

While things shake out, Flash providers are remaining optimistic. The need for what IBM may offer may not be imminent says Gordon Haff, principal IT advisor at the research firm Illuminata, “In general the need for capacity in portable devices is not growing at a very rapid rate, whereas Flash memory capacity is.” SanDisk CFO Judy Bruner, sees a pretty clear future, “We think that this marketplace for NAND Flash will continue to be a fast-growing market over the next three to five years,” and further, “We’ve been growing over the last several years at a very rapid rate. We think this is one of the more exciting marketplaces.” Scott Nelson, director of memory marketing for Toshiba America Electronic Components agrees, “The opportunity out there for new NAND products is tremendous." These two are not just talking either, their companies are backing up the words with major investments. SanDisk and Toshiba, Flash memory's original developer, recently joined hands, agreeing to invest $2.6 billion through 2008 on a Flash memory fabrication plant in Japan. The Semiconductor Industry Association (SIA) is projecting growth in the market as well, predicting that through 2009 sales of NAND Flash memory will grow at a compound annual rate of 11 percent. According to a piece in Red Herring, "In 2006 the market is projected to reach $16 billion to $17 billion, and by 2010 the market is expected to reach $34 billion."

A more complete version of this posting, with journal articles, and research reports can be found at the website of Analyst Views Weekly.

More information on this topic can be found in the Processors & Semiconductors section of Northern Light's Software, Computers, & Services Market Intelligence Center.

And in the following articles:

Flash, Meet Phase-Change Memory
Top Tech News, December 12, 2006
IBM's new phase-change memory alloy, also called GS for the two elements -- germanium and antimony -- that make it up, can quickly move between amorphous and crystalline states, reproducing the ones and zeroes used to store the world's electronic data. The GS alloy is faster than current flash memory by a factor of 500, uses half the power, and, like flash, retains its data when the power is off.

Collaborative Research Project Yields Potential Successor to Flash Memory Chips
electronic design, December 12, 2006
Working together at IBM Research labs on both U.S. coasts, the scientists designed, built, and demonstrated a prototype phase-change memory device that switched more than 500 times faster than flash while using less than one-half the power to write data into a cell. The device's cross section is a minuscule 3 by 20 nm, far smaller than flash can be built today and equivalent to the industry's chip-making capabilities targeted for 2015.

New Memory Device Could Trash Flash
Unstrung, December 11, 2006
Leap-frogging Moore's Law, scientists from IBM Corp. (NYSE: IBM - message board) will announce on Wednesday a prototype of a new type of memory device that has the potential to replace flash memory in mobile devices such as music players, cell phones, and digital cameras.

IBM Touts Flash-Killer
TechWorld, December 11, 2006
IBM has built a prototype storage device with two partners that they claim is 500 times faster than Flash. It uses less than half the power of Flash memory and can be built in ultra-thin form factors most likely unavailable to Flash. In short, a Flash-killer and potentially the answer for a universal memory type for mobile devices.

'Phase' Memory Beats Flash
PC World, December 11, 2006
Flash memory and hard-disk drives could face a challenge from a new chip technology, dubbed "phase-change" memory, being developed by a group of companies led by IBM. The companies today announced results of their latest research into the technology, which they say does a better job of storing songs, pictures and other data on iPods and digital cameras than current flash memory, and could someday replace disk drives.

Alloy Holds Out Promise of Speedier Memory Chip
New York Times, December 4, 2006
Scientists at I.B.M. and two partner companies have developed a promising material that they believe will lead to a new kind of computer memory chip able to meet the growing appetite for storing digital music, pictures and video.