Tor Aamodt

Tor Aamodt is a Professor in the Department of Electrical and Computer Engineering at The University of British Columbia. His research focuses on computer architecture, including accelerators for deep neural networks and the architecture of graphics processor units for non-graphics computing. He has contributed to the field through his work on general-purpose GPU architecture and energy-efficient computing. Aamodt has been a faculty member since 2006 and is affiliated with the university's Electrical and Computer Engineering department. His professional activities include serving on conference technical program committees and holding editorial roles. He is also recognized for his contributions to the field through various awards and recognitions.

Tor M. Aamodt is a Professor in the Department of Electrical and Computer Engineering at the University of British Columbia, specializing in Computer Architecture. His research focuses on optimizing the performance and energy efficiency of graphics processing units (GPUs) and deep learning accelerators. Aamodt's work explores advanced GPU architectures, including dynamic warp scheduling, memory access optimization, and power modeling. He has developed simulation frameworks like Accel-Sim and AccelWattch to enable detailed analysis of GPU workloads. His research also addresses challenges in deep neural network computing, such as reduced-precision strategies and bit-serial processing. Aamodt's contributions span both general-purpose GPU architectures and specialized hardware for machine learning applications.