Progress to Large Scale Memristor Neuromorphic Computing

1. A memristor has been proposed as an artificial synapse for emerging neuromorphic computing applications. (Nature Nanotechnology) To train a neural network in memristor arrays, changes in weight values in the form of device conductance should be distinct and uniform. An electrochemical metallization (ECM) memory typically based on silicon (Si), has demonstrated a good analogue switching capability owing to the high mobility of metal ions in the Si switching medium. However, the large stochasticity of the ion movement results in switching variability.

Researchers have demonstrated a Silicon memristor with alloyed conduction channels that shows a stable and controllable device operation, which enables the large-scale implementation of crossbar arrays. The conduction channel is formed by conventional silver (Ag) as a primary mobile metal alloyed with silicidable copper (Cu) that stabilizes switching. In an optimal alloying ratio, Cu effectively regulates the Ag movement, which contributes to a substantial improvement in the spatial/temporal switching uniformity, a stable data retention over a large conductance range and a substantially enhanced programmed symmetry in analogue conductance states. This alloyed memristor allows the fabrication of large-scale crossbar arrays that feature a high device yield and accurate analogue programming capability. Thus, our discovery of an alloyed memristor is a key step paving the way beyond von Neumann computing.

2. Nature Nanotechnology – Simultaneous electrodeposition of metals facilitates the realization of memrisitive devices with high yield and improved reliability.Memristors with alloyed electrodes

In 2018, memristor crossbars were made with a
single-layer density up to 4.5 terabits per inch square which was an order of magnitude denser than the state-of-the-art 64-layer triple-level cell NAND flash technology. The memristors in the crossbars are 2 × 2 square nanometers in size, capable of switching with tens of nano ampere electric current. The densely packed memristor crossbars of extremely small working devices provides a power-efficient solution for high density information storage and processing.

3 thoughts on “Progress to Large Scale Memristor Neuromorphic Computing”

  1. I don’t think we need to go neuromorphic to gain the energy efficiency of memristor computing. The early experiments done by HP with “the Machine” project pointed at an order of magnitude improvement just by avoiding copying data between different storage tiers (L1, L2, L3, RAM etc.) in a classic architecture. We love and need our algorithmic computers for their deterministic behavior. Neuromorphic computing and AI is another domain.

  2. With the death of Moore’s Law we are effectively gated by heat and energy efficiency – neuromorphic computing is really the only way forward to making a computer that could approach a mammal brain level of processing on a reasonable power budget.

Comments are closed.