This CAREER project intends to develop high-energy-efficiency computing systems by making a more "useful" elementary device rather than only focusing on its performances. This enables addressing the critical scientific question of "How can we keep pushing computing performance limits"?. For more than four decades, the semi-conductor ecosystem answered the demand for higher levels of performance by manufacturing devices with increasingly small dimensions. Nevertheless, there is still the largely unexplored route of increasing the basic switching primitive of the elementary transistors, i.e., enhancing their functionality rather than focusing only on reducing their size and/or improving their performances. This project is also relevant from an industrial perspective, as it proposes novel solutions to push device and systems performance without overextending investments to reach an ever-larger degree of integration. More importantly, this CAREER project (i) will involve graduate/undergraduate students tackling research on problems that are directly relevant for industry, ultimately boosting their employability, and (ii) will develop a scientific popularization YouTube channel as a mechanism to increase interest and broaden the participation of K-12 students by capitalizing on their online curiosity.
The proposed research aims to developing novel computing systems exploiting Three-Independent-Gate Field Effect Transistors (TIGFETs), a novel device technology capable of device-level functionalities typically not achievable by standard CMOS and leading to major benefits at gate and circuit levels. A TIGFET can, depending on its usage, achieve three totally unique modes of operations: (i) the dynamic reconfiguration of the device polarity; (ii) the dynamic control of the threshold voltage; and (iii) the dynamic control of the subthreshold slope beyond the thermal limit. In order to fully assess the potential of this technology, this CAREER project will (1) fabricate, characterize and model TIGFETs using advanced semi-conductor materials, (2) develop a complete design framework for TIGFETs, including a design kit, novel circuit primitives and dedicated design tools, and (3) evaluate TIGFET-enabled low-energy high-precision neural network computing kernels.
