A Revolution in Computing: Next–Generation Arithmetic

Wednesday, 09 Apr 2025 at 6:00 pm – South Ballroom, Memorial Union

For 50 years, "floating-point operations per second" (FLOPS) has been the currency of technical computing performance. The rise of AI and the end of Moore's law have made us realize that IEEE standard floating-point (invented by Intel in 1977) is long overdue for replacement. Dr. John Gustafson will present a new way to represent real numbers on computers that is both mathematically sound and follows engineering design goals. The approach can more than double speed and energy efficiency for everything from Machine Learning to Computer Graphics to High-Performance Computing. This is a watershed, a revolution. And it is well underway.John Gustafson is an American computer scientist and businessman, renowned for his contributions to high-performance computing (HPC). He is best known for Gustafson's Law, which provides a model for predicting the performance of parallel computing systems.Gustafson was raised in Des Moines, Iowa, and pursued his passion for science from a young age. He earned his degree in Applied Mathematics from the California Institute of Technology in 1977, followed by an M.S. and Ph.D. from Iowa State University in 1981 and 1982, respectively.Throughout his career, Gustafson has held several prominent positions, including Chief Technology Officer at Ceranovo, Inc., Chief Graphics Product Architect and Senior Fellow at AMD, and Architect of Intel Labs-SC. He also led the reconstruction of the Atanasoff–Berry computer and invented the unum number format, a variable-precision number system.Gustafson has received numerous awards for his work, including the inaugural Gordon Bell Prize in 1988 and the International Atanasoff Award in 2006. His contributions have significantly advanced the field of high-performance computing, making him a key figure in the industry.
For 50 years, "floating-point operations per second" (FLOPS) has been the currency of technical computing performance. The rise of AI and the end of Moore's law have made us realize that IEEE standard floating-point (invented by Intel in 1977) is long overdue for replacement. Dr. John Gustafson will present a new way to represent real numbers on computers that is both mathematically sound and follows engineering design goals. The approach can more than double speed and energy efficiency for everything from Machine Learning to Computer Graphics to High-Performance Computing. This is a watershed, a revolution. And it is well underway.John Gustafson is an American computer scientist and businessman, renowned for his contributions to high-performance computing (HPC). He is best known for Gustafson's Law, which provides a model for predicting the performance of parallel computing systems.Gustafson was raised in Des Moines, Iowa, and pursued his passion for science from a young age. He earned his degree in Applied Mathematics from the California Institute of Technology in 1977, followed by an M.S. and Ph.D. from Iowa State University in 1981 and 1982, respectively.Throughout his career, Gustafson has held several prominent positions, including Chief Technology Officer at Ceranovo, Inc., Chief Graphics Product Architect and Senior Fellow at AMD, and Architect of Intel Labs-SC. He also led the reconstruction of the Atanasoff–Berry computer and invented the unum number format, a variable-precision number system.Gustafson has received numerous awards for his work, including the inaugural Gordon Bell Prize in 1988 and the International Atanasoff Award in 2006. His contributions have significantly advanced the field of high-performance computing, making him a key figure in the industry.

Stay for the entire event, including the brief question-and-answer session that follows the formal presentation. Most events run 75 minutes.

Sign-ins are after the event concludes. For lectures in the Memorial Union, go to the information desk in the Main Lounge. In other academic buildings, look for signage outside the auditorium.

Lecture Etiquette

  • Stay for the entire lecture and the brief audience Q&A. If a student needs to leave early, he or she should sit near the back and exit discreetly.
  • Do not bring food or uncovered drinks into the lecture.
  • Check with Lectures staff before taking photographs or recording any portion of the event. There are often restrictions. Cell phones, tablets and laptops may be used to take notes or for class assignments.
  • Keep questions or comments brief and concise to allow as many as possible.