Computing power is an interdisciplinary topic relevant for many STEM research areas due to the need for computer visualization or simulation of physical phenomena as well as the use of machine learning in research. The Society of Physics Students is hosting Hank Childs THIS WEDNESDAY to speak on this interdisciplinary subject! This is an opportunity for CIS undergraduates will gain insight into a potential area of research and to talk with Professor Childs about one of his research topics. See the following abstract for more information. As always, we will provide pizza, snacks, and drinks — we hope to see you there!!
A typical desktop computer is approximately one million times more powerful than the first computers of sixty years ago. Further, the most powerful machines in the world are well over one million times more powerful than a desktop computer. With each increase in computing power, computers become capable of doing tasks better and better and also become capable of taking on tasks that they previously could not. When is enough? Do we need even more computing power? Or have computers become so powerful that the only limiting factor is human imagination?
In this talk, Dr. Childs will discuss several aspects of computing power. The talk will begin with a brief summary of how computers work, how we measure the power of a given computer, and how computing power has increased over time. He will then describe some applications that require computing power that challenge both today's desktop machines (computer graphics) and fastest machines (physics simulations). Dr. Childs will conclude the talk discussing the latest trends in computer hardware, and how the recent leap in computing power has enabled a corresponding leap in activities like machine learning.
Hank Childs is an Associate Professor in the Computer and Information Science Department at the University of Oregon. He received his Ph.D. in computer science from the University of California at Davis in 2006. Hank's research focuses on scientific visualization, high performance computing, and the intersection of the two. Outside of his research, Hank is best known for his role in the VisIt project, a visualization application for very large data that is used around the world. Before joining UO, Hank spent over a dozen years working for the Department
of Energy in big data visualization.