Researchers at International Business Machines (NYSE:IBM) announced a breakthrough related to phase-change memory technology on May 17. Phase-change memory is non-volatile like flash memory, meaning it continues to store data without receiving power, but it provides better performance and endurance.
Specifically, IBM researchers found a way to reliably store three bits of data per cell, up from previously being able to store just one bit per cell. According to Dr. eHaris Pozidis, manager of non-volatile memory research at IBM Research, Zurich, this progress is a big deal. "Reaching 3 bits per cell is a significant milestone because at this density the cost of PCM will be significantly less than DRAM and closer to flash."
IBM's phase-change memory is not a commercial product at this point, and no timeline was given by the company for its potential release as such. Phase-change memory could eventually be used in mobile devices, potentially replacing both DRAM and NAND. In the data center, phase-change memory could be used to store databases, boosting performance compared to flash memory and lowering cost compared to DRAM.
Does it matter?
IBM is not the first company to talk about new types of memory. Last year, Intel (NASDAQ:INTC) in partnership with Micron announced 3D XPoint, a type of memory that is both non-volatile and far faster than flash. Intel hasn't disclosed exactly what technology underpins 3D XPoint, but unlike IBM's phase-change memory, the first products from Intel aren't too far away. Intel has demoed solid-state drives under the Optane brand built with 3D XPoint, and the company expects to launch those drives sometime this year.
3D XPoint is one pillar of Intel's new strategy that focuses on the cloud. Intel already has a dominant share of the server chip market, and 3D XPoint will allow the company to increase the number of products it's selling into the data center. IBM is a threat with its OpenPOWER initiative, which now has around 150 members working to make the company's POWER chips a viable alternative to Intel's Xeon processors. If IBM succeeds in commercializing phase-change memory, it would certainly aid that effort.
IBM may still be years away from launching a phase-change memory product, so this breakthrough doesn't have any immediate ramifications for the company. But there's a good chance Intel's 3D XPoint memory won't remain unopposed for very long.