Phase-change random access memory (PRAM) is a new form of nonvolatile memory based on using electrical charges to change areas on a glassy material from crystalline to random. PRAM promises, in time, to be faster and cheaper, and consume less power, than other forms of memory.
There's a new contender coming to the realm of nonvolatile memory and storage, which enables data to remain intact when the power has been shut off.
For decades, the chief medium here has been magnetic disk. But as computers get smaller and require more and quicker storage, disk drives are lagging behind in satisfying many users??? needs.
The most recent technology to gain widespread acceptance is flash memory. USB flash drives and memory cards the size of a thumbnail that can hold several gigabytes have become important, especially for newer multimegapixel digital cameras. In 2005, consumers worldwide bought nearly $12billion worth of flash products, and the market should top $20 billion this year.
But as storage and speed requirements increase, seemingly with each new product generation, flash memory is reaching the end of its ability to keep pace. The technology can scale up only so far as the processes used to make these chips reach both practical and theoretical limits.
The new kid on the block is another solid-state technology, phase-change random access memory. Known as PRAM or PCM, it uses a medium called chalcogenide, a glassy substance containing sulphur, selenium or tellurium. These silvery semiconductors, as soft as lead, have the unique property that their physical state (i.e., the arrangement of their atoms) can be changed from crystalline to amorphous through the application of heat. The two states have very different electrical resistance properties that can be easily measured, making chalcogenide ideal for data storage.
PRAM is not the first use of chalcogenide for storage. The same material is used in rewritable optical media (CD-RW and DVD-RW), in which a laser heats up a small spot on the disk's inner layer to between 300 and 600 degrees Celsius for an instant. That alters the arrangement of atoms in that spot and changes the material's refractive index in a way that can be optically measured.
PRAM uses electrical current instead of laser light to trigger the structural change. An electrical charge just a few nanoseconds in duration melts the chalcogenide in a given spot; when the charge ends, the spot's temperature drops so quickly that the disorganized atoms freeze in place before they can rearrange themselves back into their regular, crystalline order.
Going in the other direction, the process applies a longer, less-intense current that warms the amorphous patch without melting it. This energizes the atoms just enough that they rearrange themselves into a crystalline lattice, which is characterized by lower energy or electrical resistance.
To read the recorded information, a probe measures the electrical resistance of the spot. The amorphous state's high resistance is read as a binary 0; the lower-resistance, crystalline state is a 1.
PRAM enables the rewriting of data without a separate erase step, giving the memory the potential to be 30 times faster than flash, but its access, or read, speeds don't yet match those of flash.
Once they do, PRAM-based end-user devices should quickly become available, including bigger and faster USB drives and solid-state disks. PRAM is also expected to last at least 10 times as long as flash, both in terms of the number of write/rewrite cycles and the length of data retention. Ultimately, PRAM speeds will match or exceed those of dynamic RAM but will be produced at lower cost and won't need DRAM's constant, power-consuming refreshing.
PRAM also holds out the possibility of newer, faster computer designs that eliminate the use of multiple tiers of system memory. PRAM is expected to substitute for flash, DRAM and static RAM, which will simplify and speed up memory processing.
A person using a computer with PRAM could turn it off and back on and pick up right where he left off -- and he could do so immediately or 10 years later. Such computers would not lose critical data in a system crash or when the power went out unexpectedly. 'Instant-on' would become a reality, and users would no longer have to wait for a system to boot up and load DRAM. PRAM memory could also significantly increase battery life for portable devices.
Interest in chalcogenide materials began with discoveries made by Stanford R. Ovshinsky of Energy Conversion Devices Inc., now known as ECD Ovonics, in Rochester Hills, Mich. His work revealed the potential for using those materials in both electronic and optical data storage. In 1966, he filed his first patent on phase-change technology.
In 1999, the company formed Ovonyx Inc. to commercialize PRAM, which it calls Ovonic Universal Memory. ECD licensed all of its intellectual property in this area to Ovonyx, which has since licensed the technology to Lockheed Martin Corp., Intel Corp., Samsung Electronics Co., IBM, Sony Corp., Matsushita Electric Industrial Co.'sPanasonic unit and others. Ovonyx's licenses center around the use of a specific alloy of germanium, antimony and tellurium.
Intel invested in Ovonyx in 2000 and 2005 and has announced a major initiative to replace certain types of flash memory with PRAM. Intel has built sample devices and plans to use PRAM to replace NAND flash. It hopes to eventually use PRAM in place of DRAM. Intel expects Moore's Law to apply to PRAM development in terms of cell capacity and speed.
As yet, no commercial PRAM products have reached the market. Commercial products are expected in 2008. Intel expects to show sample devices this year, and last fall Samsung Electronics showed a 512Mbit working prototype. In addition, BAE Systems has introduced a radiation-hardened chip, which it calls C-RAM, intended for use in outer space.
Kay is a Computerworld contributing writer in Worcester, Mass. You can contact him at firstname.lastname@example.org.