With the complexity of computing tasks and the increase of data volume, the traditional general computing platform has been unable to meet the needs of high performance computing. With the acceleration of heterogeneous computing systems, different computing platforms have different instruction sets and architecture characteristics. The compiler architecture for heterogeneous computing can provide higher performance and efficiency, and support seamless integration between different types of computing units and platforms, thus promoting the development and innovation of computing technology.

Speakers:

Wang Chenhan: OpenBayes Bayesian computation, Founder and CEO of OpenBayes, He has worked for world famous companies such as Walt Disney Interactive Media Group and AVOS Systems, Secretary general of CLUE Foundation, a benchmark for Chinese semantic understanding, and deputy director of the Bayesian Computing Joint Research Center of Tianjin University.