Skip to main content

Topics in Physics (390-0-1)

Instructors

Kristian Allan Hahn

Meeting Info

Technological Institute L168: Tues, Thurs 12:30PM - 1:50PM

Overview of class

The field of high-performance scientific computing is increasingly leveraging heterogeneous system architectures to improve the execution of demanding workflows. Heterogeneous architectures combine CPUs with additional hardware accelerators, the most familiar of which is perhaps the general purpose GPU. A less familiar but often more powerful alternative is the Field Programmable Gate Array (FPGA), which can provide reduced latency, higher throughput, greater flexibility, and significant power savings. In this course we will consider Monte Carlo integration as a case study for an exploring heterogeneous computing with FPGAs.

Students should have a working knowledge of basic C/C++. We will briefly review the relevant aspects of the language(s) at the beginning of the course, but students will greatly benefit from having had a more formal, earlier introduction. No specific C/C++ textbook is required. Some prior exposure to introductory statistics and experience with Unix/Linux operating systems will be helpful, although this is also not required.