MIT, Stanford Researchers Demo 3D Computing Chips Capable of Storage and Computing

• By Joshua Bolkan
• 07/10/17

Researchers at the Massachusetts Institute of Technology (MIT) and Stanford have created a prototype 3D computer chip that combines computing and data storage.

Traditional computer chips are two-dimensional because creating silicon chips requires temperatures of more than 1,000 degrees Celsius and the heat of layering a second chip on top of another would damage the existing chip.

But that flat design has limitations. Computing chips and data storage chips are separate and there are limited connections between them, creating a bottleneck for applications that analyze massive amounts of data.

The new chips, developed by a team led by Max Shulaker, assistant professor of electrical engineering and computer science at MIT, use carbon nanotubes instead of silicone to build chips that can be created at temperatures below 200 degrees Celsius and can thus be stacked.

The new chips use resistive random-access memory (RRAM) and alternate layers of storage and computing chips with ultra-dense wires between them in an effort to address the data bottleneck.

"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," said Subhasish Mitra, professor of electrical engineering and computer science at Stanford and Shulaker's adviser when he was a PhD candidate there, according to a report from MIT. "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."

"It leads to a fundamentally different perspective on computing architectures, enabling an intimate interweaving of memory and logic," said Jan Rabaey, a professor of electrical engineering and computer science at the University of California at Berkeley who was not involved with the research, according to MIT News. "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."

As a demonstration of the technology, Shulaker and his team added a layer of nanotubes to the top of the chip with more than 1 million sensors capable of detecting ambient gasses in the air.

"Due to the layering of sensing, data storage, and computing, the chip was able to measure each of the sensors in parallel, and then write directly into its memory, generating huge bandwidth, Shulaker says," according to information released by MIT.

A paper about the chip was published in the most recent issue of nature and is available for a fee at nature.com. More information is available at news.mit.edu.