Towards a science of high performance design
Assistant Research Professor
Department of Electrical and Computer Engineering
Carnegie Mellon University
20 Mar 2018 Tuesday, 02:00 PM to 04:00 PM
COM2 Level 4
Executive Classroom, COM2-04-02
Applications are becoming more varied. Architectures are becoming more complex. Yet software implemented by the expert still achieves higher performance than most automatically generated code despite two decades of research in automatic empirical optimization systems. Can we ever get expert-level performance automatically? In this talk, I will discuss how expert-level performance in the dense linear algebra domain can be systematically attained through the use of formal methods and analytical models. Specifically, I will present our analytical models and their underlying hardware principles. I will also share how analytical models from the dense linear algebra domain have been adapted to design high performance implementations for problems in other domains.
Tze Meng Low is an Assistant Research Professor with the Department of Electrical and Computer Engineering at Carnegie Mellon University. He graduated from the University of Texas at Austin with a Ph.D. in Computer Science in 2013. His research focuses on the use of formal methods and analytical models to achieve the vertical integration of high performance algorithms, software, and hardware. Currently, he is looking to extend the models and insights from high performance dense linear algebra to apply them to other domains such as bioinformatics, signal processing, deep learning and graph algorithms.