The development of a universal technology that replaces current memory types has long been a major objective in the semiconductor market. Experts have predicted that the advent of a one-size-fits-all memory would overturn the order of the more than $200 billion industry, substituting all entrenched product categories with a single device. However, major new technological developments often don’t fit neatly into the current order of things, and the impact of such a device on the technology market could transcend the memory business.
A universal solution would combine the best aspects of current dominant memory types—including the speed of SRAM, the non-volatility of flash and the low cost and high density of DRAM. Over the years, multiple technologies have been proffered as candidates for the universal memory, including phase-change memory, magnetoresistive random-access memory (MRAM), ferroelectric RAM (FRAM) and carbon-nanotube-based RAM.
In 2005, market research firm iSuppli—now IHS Inc.—forecasted that a universal device could eventually account for 80 percent of all memory market revenue. This would amount to $168 billion in a total memory market worth more than $210 billion this year, according to IHS.
The arrival of such a device would bring major technological and market disruption to a well-established industry dominated by major suppliers and product segments closely tied to high-volume markets.
However, the impact of universal memory could be much greater than that. For example, in the 1990s, many technology experts predicted that flash memory’s destiny was to supplant the hard-disk drive (HDD) as the storage system in personal computers. However, while flash-based solid-state drives still haven’t overtaken HDDs in PCs, the memory technology helped usher in a much more fundamental change to the technology industry. Flash serves as the enabling storage medium for a new generation of computing devices—smartphones and media tablets. These platforms have overtaken personal computers in terms of shipments and have led to the post-PC era.
The potential impact of a universal memory may be evident in the development of a new product called the memristor. Now in development by Hewlett-Packard Co., the memristor remembers the level of resistance that naturally occurs in circuits in order to store information.
Reports from HP indicate that memristor technology fits the bill for the universal technology, with speed, non-volatility and—eventually—low cost. HP believes that the devices could serve as both system memory and storage, replacing DRAM and flash memory—or HDDs—in computer systems.
HP believes memristor technology eventually could lead to the creation of an entirely new type of computer that mimics the functionality of the human brain, recognizing patterns. The development of a computer with the capabilities of the human brain would represent one of the greatest technological achievements of all time—and needless to say, would have much greater impact on society than a shakeup of the memory chip market.
HP’s ambitions for memristors demonstrate the huge impact that an advanced universal technology could have beyond the confines of the memory market.