Carbon nanotube NRAM computer memory will start to ship in 2019

Nantero is developing next-generation memory using carbon nanotubes. NRAM’s has DRAM-like speed, nonvolatility and low cost.

Nantero and its partner Fujitsu aims to ship a DDR4 DRAM alternative using the technology in 2019.

Nantero’s non-volatile NRAMs use electrostatic charge to activate systolic arrays of CNT cells it claims are relatively easy to sputter on to any CMOS process. It claims it will outstrip the DRAM roadmap starting with stacks of 4-Gbit, 100mm2 die made in a 28nm process into initial 8- and 16-Gbit chips.

DRAM is expected to hit a wall around the 64-Gbit device. Micron is exploring alternatives such as phase-change memories.

Nantero forecasts a 64 Gbit NRAM could be made in a 14nm process and a 256-Gbit device in 7nm, both using four layers.

Nantero created a DDR4 reference design “to get the technology moving…this is scalable way beyond DRAM, so I think we have a comfortable road map to replace DRAM,” Gervasi said.

* A multi-Gb DDR4-compatible nonvolatile standalone memory product with speed comparable to DRAM and a lower price/GB than DRAM
* A standalone chip designed as a cache for SSDs or HDDs which removes the need for battery backup and allows dramatic expansion of cache size, substantially speeding up the drive
* A variety of embedded memory offerings, including a highly scalable embedded nonvolatile memory that can scale to 5nm in size and operate at DRAM-like speeds, and operates under very high temperature, making it ideal for automotive applications, as well as IOT

6 thoughts on “Carbon nanotube NRAM computer memory will start to ship in 2019”

  1. I have been hearing about NRAM for about.. a decade ? now. Wondering if it still makes sense in a sub 10nm world

  2. I have been hearing about NRAM for about.. a decade ? now. Wondering if it still makes sense in a sub 10nm world

  3. Getting rid of the need to refresh would still be a big deal for power consumption and performance, even if it doesn’t improve density.

  4. Getting rid of the need to refresh would still be a big deal for power consumption and performance even if it doesn’t improve density.

Comments are closed.