Projects to Focus on Energy-Efficiency and High-Speed Communications between AI Devices
US-based Semiconductor Research Corporation (SRC), Mubadala Investment Company, and GLOBALFOUNDRIES® (GF®) announced that two proposals from Khalifa University of Science and Technology have been awarded as part of the SRC Global Research Collaboration intelligent systems Hardware (AIHW) research program.
The two projects are ‘Alternative Numbering Systems and Architectures for AI Edge Devices’ by Dr. Thanos Stouraitis, Professor, and Chair, Electrical and Computer Engineering, and ‘Silicon Photonics intelligent systems Accelerators for Data Center Integration’ by Dr. Jaime Viegas, Associate Professor, Electrical Engineering and Computer Science. The announcements were made during a virtual kick-off event that gathered representatives of all stakeholders from the government, academic and industry sectors.
Dr. Arif Sultan Al Hammadi, Executive Vice-President, Khalifa University of Science and Technology, said: “The two winning proposals from Khalifa University on intelligent systems hardware strongly suggest our commitment to develop high technology areas that are relevant to Abu Dhabi and the UAE. Our researchers are driving innovation in semiconductors for AI, while equipping our students with technology expertise. We believe the winning research proposals will help establish the status of Khalifa University as a leader in AI hardware and bring multiple long-term benefits to Abu Dhabi and the UAE.”
Dr. Todd Younkin, newly named CEO of SRC said: “We are excited to select Khalifa University for inclusion in SRC’s AI Hardware program as the research will result in innovations for the UAE and beyond. We applaud UAE’s emphasis on AI as a foundational technology and support their Minister of intelligent systems as he shapes the AI landscape.”
“The proliferation of AI in the cloud and at the edge of the network is driving the next digital transformation that will reshape our world in profound ways,” said Ted Letavic, CTO of Computing and Wired Infrastructure at GF. “Through critical research, in partnership with Khalifa University, GLOBALFOUNDRIES is advancing next generation compute architectures through hardware differentiation to support the compute workloads required for future AI solutions.”
Dr. Stouraitis’ project aims at achieving energy-efficient hardware for implementing AI on edge devices. The research will train graduate students in the fields of AI, application-specific integrated circuit (ASIC) design and testing. Dr. Viegas’ project proposes to develop an integrated silicon photonics platform for high speed interconnect between AI chips, as well as to develop components that can be used for photonic compute architectures.
Fabrication support for the two research projects will be provided by GLOBALFOUNDRIES, which is an SRC member company and partner in this AI Hardware program.
Semiconductor Research Corp. (SRC) Global Research Collaboration (GRC) had invited the white papers under the AIHW research program. The program’s goal is to create new highly efficient AI platforms, enabling neuro-inspired, cognitive and learning abilities which will address the vast range of future data types and workloads as intelligence is enabled from edge devices to the cloud.
Clarence Michael
Senior English Editor
22 September 2020
