Smartphones are getting increasingly feature-rich and with that, the need for high performance chipsets is also increasing. For processors to perform at par with evolving smartphone technology such as virtual reality and artificial intelligence, they will need to sort bandwidth issues between the processor and RAM.
A research team from the Massachusetts Institute of Technology (MIT) and Stanford University might have a solution. It has developed a prototype processor made of Graphene carbon nanotubes, with a resistive RAM layer on top. The prototype effectively combines RAM and CPU, which will not only save space in a smartphone motherboard, it will also create better performance. The model could not be implemented on traditional silicon-based chipsets, which would heat up and damage the RAM.
“Circuits today are 2-D, since building conventional silicon transistors involves extremely high temperatures of over 1,000 degrees Celsius,” says Max Shulaker, a core member of the research team. “If you then build a second layer of silicon circuits on top, that high temperature will damage the bottom layer of circuits.”
Instead, the team used 3D architecture — a computerized design which interweaves logic and memory on a chipset. The researchers integrated 1 million RAM cells and 2 million carbon nanotube-based transistors into a chipset, which it calls “the most complex nanoelectronic system ever made.”
Carbon nanotubes enable the RAM to be fabricated at much lower temperatures, below 200 degree centigrade, which is generally needed by silicon based chipset designs.
According to Subhasish Mitra, advising professor on the project, the new design will also make for better performing chipsets.
“The new 3-D computer architecture provides dense and fine-grained integration of computating and data storage, drastically overcoming the bottleneck from moving data between chips. As a result, the chip is able to store massive amounts of data and perform on-chip processing to transform a data deluge into useful information,” Mitra said.
The carbon nanotubes will also double up as sensors, which will detect and classify ambient gases. These sensors will then send signals directly to the RAM, which will result in a higher signal bandwidth. The new processor system will be suited for artificial intelligence and deep learning based tasks. It would also be able to perform tasks that current processors can’t, such as detecting signs of disease by sensing particular compounds in a patient’s breath.
“It leads to a fundamentally different perspective on computing architectures, enabling an intimate interweaving of memory and logic. These structures may be particularly suited for alternative learning-based computational paradigms such as brain-inspired systems and deep neural nets, and the approach presented by the authors is definitely a great first step in that direction,” Jon Rabaey, a professor of electrical engineering and computer science at the University of California at Berkeley, who was not involved in the research, was quoted saying in the press release from MIT.
The team is now exploring the production of the chipset, along with a company called Analog devices.