2023 Keynote Speakers

On the Advance of Moore’s Law and Resulting Trends in Reliability

Ann Kelleher (Intel)

Abstract: The semiconductor industry is driving innovation at a Moore’s Law cadence to satisfy the world’s ever-increasing demand for compute and connectivity. Innovation is driven by customer needs for product functionality, and is enabled by novel transistor, interconnect, memory, and packaging architectures. These innovations across the industry increase system complexity and impact reliability. This keynote discusses future silicon, packaging technologies and reflects upon their implications for the reliability community. The benefits of a System Technology Co-Optimization (STCO) approach for reliably delivering functionality will also be briefly discussed.

Dr. Ann B. Kelleher is executive vice president and general manager of Technology Development at Intel Corporation. She is responsible for the research, development and deployment of next-generation silicon logic, packaging and test technologies that power the future of Intel’s innovation. Previously, Kelleher was general manager of Manufacturing and Operations, where she oversaw Intel’s worldwide manufacturing operations including Fab Sort Manufacturing, Assembly Test Manufacturing and strategic planning, as well as corporate quality assurance and corporate services. Before that, she served as co-general manager of the Technology and Manufacturing Group. Kelleher joined Intel in 1996 as a process engineer, going on to manage technology transfers and factory ramp-ups in a variety of positions spanning 200mm and 300mm technologies. She started her manufacturing leadership journey as the factory manager of Fab 24 in Leixlip, Ireland. She then moved to Chandler, Arizona, to manage Intel’s Fab 12 facility, followed by serving as the site manager of Intel’s Fab 11X fabrication facility in Rio Rancho, New Mexico. She then became general manager of the Fab Sort Manufacturing organization where she was responsible for all aspects of Intel’s high-volume silicon manufacturing. Kelleher holds a bachelor’s degree, a master’s degree and a Ph.D. in electrical engineering, all from University College Cork in Ireland.

Title: Transforming Industries with Trustworthy Cloud-to-Edge Compute Platforms

Gary Hicok (NVIDIA)

Abstract: Advanced computing systems are beginning to be used in critical life-saving applications. High-performance solutions are transforming transportation, medical and other vital information technology systems, along with drug discovery, climate research and more. Trusting these systems, which are composed of silicon, hardware, software and reliant on a huge amount of data, has become essential to success. This trust is primarily built on reliability, safety and security.

Along with the rising need for trustworthiness, platforms themselves are advancing, becoming more complex and scalable. They extend from the cloud to high-performance embedded systems with hybrid systems-of-systems that feature parts of each. This new unit of computing opens up more potential for failure, but also affords more opportunities to improve trust by cross-utilizing hardware, software and system techniques. This talk covers cross-discipline and cross-system approaches to check, monitor, predict, protect, deliver redundancy and simulate end-to-end compute platform trustworthiness.

Gary Hicok
Gary Hicok is senior vice president at NVIDIA and is responsible for Product Engineering, which oversees Shield, Jetson, Robotics and DRIVE platforms. Prior to this role, Hicok served as senior vice president of NVIDIA’s Mobile Business Unit. This vertical focused on NVIDIA’s Tegra mobile processor which was used to power next-generation mobile devices as well as in-car safety and infotainment systems. Before that, Hicok ran NVIDIA’s Core Logic (MCP) Business Unit also as senior vice president.  Throughout his tenure with NVIDIA, Hicok has also held a variety of management roles since joining the company in 1999 with responsibilities focused on console gaming and chipset engineering. He holds a BSEE degree from Arizona State University and has authored 38 issued patents.

Reliability Challenges for the Next Decade of High-Performance Computing

Mark Fuselier (AMD)

Abstract: As classic Moore’s law has slowed to a crawl, our industry faces many challenges as we drive the continued scaling of High-Performance Compute solutions. Our industry is spearheading innovation across silicon technology, chiplet integrated advanced packages, high bandwidth I/O, and custom heterogenous compute architectures. These advancements bring new and distinct quality, reliability, and physical failure analysis issues that we must tackle. This keynote will call our community to action while it addresses key innovations necessary over the next decade to push the performance scaling required to support the compute bandwidth required to solve our harshest challenges.

Mark Fuselier is Senior Vice President of Technology and Product Engineering at AMD. He is responsible for end-to-end foundry technology and engineering for new product introduction and manufacturing.  Fuselier has more than 25 years of semiconductor industry experience and has been involved in the development and production of process technology generations spanning from 0.35um through 7nm across multiple fabs and products. He has played a central role in the development and productization of system-level solutions such as multi-core CPU SoC integration, heterogenous APUs, 2.5D chip-package, and chiplet System in Package (SiP) integration.  Fuselier holds a Master of Science degree in Electrical Engineering and Master of Business Administration from the University of Texas at Austin. He is a member of IEEE and the Electron Devices Society

Building Reliability into the Modern Cloud

Rohit Vidwans (Ampere)

Abstract: As modern workloads move to the cloud at an unprecedented speed, Ampere’s 80-core Ampere® Altra® and the 128-core Ampere® Altra® Max Cloud Native Processors are disrupting the status quo of legacy technology. Ampere is implementing a novel design methodology and state-of-the-art architecture to deliver reliable, high-quality, and consistent workload execution at the lowest possible power consumption and longer life expectancy. In his talk, Rohit will discuss the importance of reliability within these power-efficient multi-core designs, share some of the extensive testing and product qualification methodologies being implemented to address unique cloud-based core-utilization models and discuss other emerging challenges that Ampere will work with the industry to resolve in future products for the Modern Cloud.

Rohit Vidwans
EVP and Chief Engineering & Manufacturing Officer for Ampere Computing, a modern semiconductor company focused on inventing the future of cloud-based processor technology. In this role, he and his experienced technical team are designing Cloud Native Processors that provide a new level of performance and efficiency to handle the large compute demands of today’s hyperscale cloud and edge computing workloads. Prior to Ampere, he worked at Intel for over 25 years where he developed microprocessors, graphics, media and supercomputing processors and platforms, and holds 8 microprocessor design patents. Notable projects include Intel’s first 8 and 10 core Xeon® microprocessors for data center and enterprise servers and Intel’s first multi-core phone and tablet SOCs based on the ATOM® core. Rohit has a MSCE from the University of Texas at Austin and a BSCE from the University of Michigan at Ann Arbor.

Archives

IRPS 2022 Keynote Speakers

IRPS 2021 Keynote Speakers