Heterogeneous Computer Architectures
Zhiru Zhang, Electrical and Computer Engineering, wants to significantly improve the design, productivity, and quality of heterogeneous computer architectures. With this National Science Foundation CAREER Award, Zhang will develop a synergistic design automation framework where a set of novel programming abstractions, architectural templates, synthesis optimization algorithms, and hardware prototypes all play concerted roles to overcome the many challenges raised by irregular programs. The goal is to automatically generate softly synthesized accelerators capable of decoupling data access from computation for tolerating memory latency and performing run-time optimizations, exploiting the irregular parallelism. The new design automation methodologies and tools will enable productive modeling, exploration, and generation of hardware accelerators relevant to computer vision, machine learning, physical simulation, and social network analytics.