This site may earn affiliate commissions from the links on this page. Terms of use.

For years, scientists and researchers have looked for the so-called Holy Grail of memory applied science — a not-volatile memory standard that'due south faster than NAND flash while offering superior longevity, higher densities, and ideally, ameliorate ability characteristics. I of the more promising technologies that's been in development is phase-change memory, or PCM. IBM researchers announced a major breakthrough in PCM this calendar week, declaring that they've establish a way to store upward to three $.25 of data per "cell" of memory. That's a significant achievement, given that previous piece of work in the field was limited to a unmarried scrap of data per memory jail cell.

Phase alter memory exploits the backdrop of a metal alloy known as chalcogenide. Applying rut to the blend changes it from an amorphous mass into a crystal lattice with significantly unlike properties, equally shown beneath:

Theseus1

Scientists take long known that chalcogenide could be in states between crystal lattice or amorphous, but building a solution that could exploit these in-betwixt states to shop more retention has been extremely hard. While phase-change memory works on very dissimilar principles than NAND flash, some of the problems with scaling NAND density are conceptually similar to those faced by PCM. Storing multiple bits of information in NAND flash is difficult because the gap betwixt the voltage levels required to read each specific chip is smaller the more $.25 you store. This is also why TLC NAND flash, which stores three $.25 of information per cell, is slower and less durable than MLC (2-bit) or SLC (single scrap) NAND.

IBM researchers take discovered how to store iii bits of data in a 64K array at elevated temperatures and for one meg endurance cycles.

"Phase change memory is the start instantiation of a universal retentiveness with properties of both DRAM and flash, thus answering one of the grand challenges of our industry," said Dr. Haris Pozidis, an author of the paper and the manager of non-volatile memory research at IBM Research – Zurich. "Reaching iii bits per cell is a significant milestone considering at this density the toll of PCM volition be significantly less than DRAM and closer to flash."

Here'due south how the PR blast describes the breakthrough:

To achieve multi-chip storage IBM scientists have developed two innovative enabling technologies: a prepare of drift-immune cell-state metrics and drift-tolerant coding and detection schemes.

More specifically, the new cell-state metrics measure a physical belongings of the PCM cell that remains stable over time, and are thus insensitive to migrate, which affects the stability of the prison cell'south conductivity with time. To provide additional robustness of the stored data in a prison cell over ambient temperature fluctuations a novel coding and detection scheme is employed. This scheme adaptively modifies the level thresholds that are used to observe the jail cell's stored information and then that they follow variations due to temperature change. As a result, the cell state tin can be read reliably over long time periods subsequently the memory is programmed, thus offering non-volatility.

"Combined these advancements address the cardinal challenges of multi-bit PCM, including drift, variability, temperature sensitivity and endurance cycling," said Dr. Evangelos Eleftheriou, IBM Fellow.

In that location'south however a nifty deal of piece of work to do before stage-alter memory tin can be considered equally a candidate to replace NAND wink or DRAM in sure situations. The operation and power impact of these new structures has non been characterized and the switching time hasn't been revealed.

Universal-PCM

The graphic above is from an IBM video explaining how PCM memory works and some general information on this latest breakthrough. Note that PCM, like NAND flash, takes a performance striking when it shifts to a multi-bit architecture. While single-chip PCM is nearly as fast as DRAM (according to IBM), multi-bit PCM is significantly slower. Information retention (how long information remains in the cell) was too worse than NAND wink, which has lower endurance (how many read/write cycles the cells can withstand) but higher data retention.

Stage-change memory is theoretically capable of replacing DRAM in at least some instances, but if these density gains come at the cost of programming speed, the net proceeds may be minimal. Phase-modify retentivity also requires big amounts of power to plan and generates a great bargain of heat every bit a outcome.

This video from IBM walks through the history of phase-change memory, explains the basics of its role, and covers the most-recent breakthrough. Nosotros think IBM's discovery here could assistance pave the way for a long-term replacement to NAND flash, but we're still years away from that. Intel'south Optane 3D XPoint retention may brand its own play for the server and data center space, and Micron, which used to manufacture PCM for the mass market place doesn't build it anymore.