Researchers at Melbourne's RMIT University have built a data storage nano structure that mimics the human brain and could open the floodgates for the development of nanoscale electronic memory devices.
Dr Sharath Sriram, project leader and co-leader of the RMIT Functional Materials and Microsystems Research Group, said that the nanometre-thick stacked structure central to the new technology was created using a film of oxide material over 10,000 times thinner than a human hair.
"The thin film is specifically designed to have defects in its chemistry to demonstrate a 'memristive' effect — where the memory element's behaviour is dependent on its past experiences," said Sriram.
From Sriram's perspective, the nano structure could be used for a range of electronic applications, from ultra-fast memory devices just a few nanometres thick to computer logic architecture that replicates the responses of a biological neural network.
Importantly, the research helps open the door to the exploration of new materials in the non-stop search for smaller, faster data memory devices.
"With flash memory rapidly approaching fundamental scaling limits, we need novel materials and architectures for creating the next generation of non-volatile memory," he said.
Industry experts havethat the physical size limit for operational NAND flash is about 10 nanometres. Technology players are getting very close to this threshold, with Toshiba — the company that first introduced flash storage — developing 15-nanometre process technology earlier this year, while maintaining the same write speed as larger 19-nanometre process technology.
The new RMIT University research, supported by the Australian Research Council Discovery grant, relies on memristors (memory resistors), which have been touted as a potential replacement for current hard drive technologies such as flash, SSD, and DRAM.
Although more work is yet to be done on the project, which is the result of a joint collaboration between members of the Functional Materials and Microsystems Research Group and professor Strukov from the University of California Santa Barbara, it is possible that the new technology could eventually provide the infrastructure for a system that mimics the human neural network.
"While more investigation needs to be done, our work advances the search for next-generation memory technology can replicate the complex functions of human neural system — bringing us one step closer to the bionic brain," he said.
The new research by RMIT University follows the unveiling in August of IBM's, SyNAPSE, which the company said functions like a brain and would open new computing possibilities for cloud, mobile, and distributed sensor applications.
At 5.4 billion transistors, IBM said the chip is currently one of the largest CMOS chips ever built, while also managing to consume only 70mW of power by running at biological real time.
On September 25, IBM Research's Dharmendra S Modha said in a column in Forbes, that the new chip could "transform science, technology, business, government, and society by enabling vision, audition, and multi-sensory applications".
Modha said that looking into the future, researchers are working on integrating multi-sensory neurosynaptic processing into mobile devices, and hope to eventually create neurosynaptic supercomputers by tiling multiple chips on a board in order to build systems that could scale to "100 trillion synapses and beyond".