Performance Engineering for Scalable AI
February 12, 2024 @ 11:00 am – 11:50 am
Speaker: Xuhao Chen, Research Scientist, MIT CSAIL
Hosted by Stevens Institute of Technology
Location: Gateway North 303 (with virtual option)
ABSTRACT
Due to massive size, complex algorithms and irregular data structures, AI applications are expensive and hard to scale, which poses great challenges in computing system design.
In this talk, I will discuss my approach called cross-stack performance engineering, to address this challenge. I will describe software and hardware system design principles, optimization techniques and automation methodologies, across different layers of the system stack. I will show promising results to demonstrate that this cross-stack approach is effective to make AI scalable.
BIOGRAPHY
Xuhao Chen is a Research Scientist at MIT CSAIL. Dr. Chen is broadly interested in parallel systems and architectures, with a focus on AI and big-data applications. His recent work aims to make AI scalable by designing efficient algorithms, software systems and hardware accelerators. His work has been published in OSDI, ISCA, MICRO, VLDB, ICS, etc.
Zoom link: https://stevens.zoom.us/j/91365277622