With the advent of multicore processors, we are faced with the challenge of exploiting this ubiquitous multi-threaded hardware. This course explores the evolution of modern parallel architectures and programming models, and is divided into two phases.

In the first phase, we will investigate in detail the design and operation of modern parallel architectures, with a brief look at how they are programmed. The second phase of the course will switch gears: we will study current research and development of emerging parallel architectures including multicore processors, interconnection networks, and accelerators and heterogeneous systems-on-chip. In this phase we will read research papers, and through the class project, implement and evaluate new ideas.

Students are welcome to suggest topics for class discussion and/or projects.