Imec has launched a consortium with 26 European university groups to work on the technology roadmap beyond CMOS scaling (CMOS 2.0). The initiative will focus on design automation and chip architecture research for the next generation of chips. The partners, including Delft University of Technology, Ghent University, KU Leuven and Université Libre de Bruxelles, will benefit from the NanoIC pilot line, turning academic insights into industry-focused innovations. In the future, similar consortia will be set up around advanced materials and alternative compute systems.
CMOS 2.0 refers to a new paradigm, introduced by Imec, that expands the chipmaking toolbox beyond traditional transistor scaling and its associated scaling challenges. It allows for more design flexibility by exploiting fine-grain wafer stacking technology to improve on-chip connectivity and offer higher technology heterogeneity to the system. It will result in tailored chips comprising multiple 3D-stacked layers that fulfil smartly partitioned functions. In that way, CMOS 2.0 will provide advanced, versatile platforms that push the boundaries of compute performance.
Introducing this new paradigm will have profound implications on how computing architectures are designed and optimized for future workloads and applications. CMOS 2.0 is a key differentiator for the realization of next-generation energy-efficient compute systems. It’s expected to impact a wide variety of applications, from general-purpose processors to high-performance AI computing systems and even further for embedded AI applications at the edge.
Within the CMOS 2.0 consortium, 26 PhDs will be funded. They’ll stay at their home university, embedded in their research group, allowing them to tap into complementary fields of expertise and stimulate cross-fertilization. The participating universities and Imec will jointly develop the necessary know-how that lays down the foundation of the next generation CMOS technology platforms and their associated compute architectures. Moreover, the collaboration will support workforce and skill development in Europe to meet current and future industry needs.
“The attraction for the concept of CMOS 2.0 is clear, but the obstacles are equally substantial,” explains Sahar Sahhaf, Imec’s director of academic partnership development. “Leveraging the benefits in both connectivity and heterogeneous integration enabled by 3D wafer stacking will reshape every stage of design and chip architecture. It requires convergence of expertise, close collaboration, and coordination. It’s the first time that Imec brings together such a network of European university teams in a structured way to have guided contributions to the future semiconductor roadmap. We’re excited to further connect academic inputs in our industry-driven programs to put Europe in the forefront of research on advanced computing technologies.”
Mehdi Tahoori, professor and chair of dependable nano-computing at Karlsruhe Institute of Technology (KIT), Germany, and technical director of the new academic collaboration, adds: “This university research consortium aims to infuse CMOS 2.0 technology into the entire design stack, from electronic design automation all the way to systems architecture. It aims to stimulate the broader research and academic community on various aspects of the CMOS 2.0 revolution.”
Of particular importance is the presence of the NanoIC pilot line hosted by Imec in Leuven and its role in empowering the CMOS 2.0 academic consortium. Its tools are embedded within a collaborative ecosystem of industry partners. PhD students can gain early exposure to next-generation semiconductor logic, memory and 3D technologies through process design kits (PDKs), which will enable them to develop system-level thinking – which is typically only encountered much later in a research or industrial career. As such, it bridges the gap from academia to industry, facilitating a rapid transfer of knowledge and advanced technology from research labs to the market, thus strengthening Europe’s industry.
Next to Imec, four universities from the Benelux and KIT, the consortium includes Greece’s National Technical University of Athens and University of Thessaly, École Polytechnique Fédérale de Lausanne and Eidgenössische Technische Hochschule Zürich from Switzerland, KTH Royal Institute of Technology in Sweden, the LIRMM joint research unit of the University of Montpellier and CNRS in France, Italy’s Politecnico di Torino, Sabancı University in Turkey and Universidad Complutense de Madrid, Spain.
