A Memory Breakthrough
Two firms have doubled the capacity of phase-change memory, a likely replacement for flash.
Over the past decade, flash memory has changed the electronics landscape, giving us robust storage in tiny devices such as iPods and cell phones. As chip sizes shrink, however, engineers know there will be limits to flash performance, and they have been eyeing a replacement technology called phase-change memory. Today, Intel announced a research advance that doubles the storage capacity of a single phase-change memory cell. This new approach is also implemented in the chip via algorithms so that it won’t add cost to the existing phase-change memory fabrication process.
Phase-change memory differs from other solid-state memory technologies such as flash and random-access memory because it doesn’t use electrons to store data. Instead, it relies on the material’s own arrangement of atoms, known as its physical state. Previously, phase-change memory was designed to take advantage of only two states: one in which atoms are loosely organized (amorphous), and another where they are rigidly structured (crystalline).
But in a paper presented at the International Solid State Circuits Conference in San Francisco, researchers illustrated that there are two more distinct states that fall between amorphous and crystalline, and that these states can be used to store data.
To make their memory cells, Intel and partner ST Microelectronics used a material called GST, a type of glass that has physical states responsive to heat. A tiny heater, controlled by algorithms in the chip, changes the state of the GST by heating a memory cell until it reaches one of four distinct states. (Older systems used the same approach but only worked with two states.) Intel’s chief technology officer, Justin Rattner, says the researchers used novel programming algorithms to alter the amount of heat each cell receives, thus controlling its state: “We can do this successfully with a reasonably sized array, and do it at speeds that are commercially viable,” he says. The cell is then read by measuring its electrical resistance between two electrodes. The resistance indicates the state of the cell because each state has distinct electrical properties.
By adding two bits per cell, Intel and ST Microelectronics have put phase-change memory on par with today’s flash technology, says H.-S. Philip Wong, professor of electrical engineering at Stanford University. Intel has already mastered a similar trick with flash memory in which more than one bit can be stored per memory cell, he says, so this is a logical progression for phase-change memory. “It’s rather important to develop this multi-bit storage technology,” says Wong. “If you can’t do it, then you’re disadvantaged by a factor of two.”
One of the features that makes phase-change so compelling as a flash alternative is that it has the same benefits as flash with faster speed, says Jim Handy, an analyst at Objective Analysis, a semiconductor market research firm. Like flash, phase-change memory is a non-volatile memory that can store bits even without a power supply. But unlike flash, data can be written to cells much faster, at rates comparable to the dynamic and static random-access memory (DRAM and SRAM) used in all computers and cell phones today. Currently, Handy explains, computer- and cell-phone engineers use DRAM or SRAM combined with flash. DRAM and SRAM are used to read and write data quickly; flash is used to store data when the power is off. “Handset manufacturers are excited about phase-change memory,” Handy says, “because it looks like they could get rid of two of the chips [flash and DRAM] and replace them with one phase-change memory chip.”
Phase-change memory has made a lot of progress in the past few years, Wong adds. “A few years ago it looked promising,” he says. “But now it’s going to happen. There’s no doubt about it.”
The AI revolution is here. Will you lead or follow?
Join us at EmTech Digital 2019.