当前位置 >>  首页 >> 合作交流 >> 学术交流

Dr. Christophe Dubach Gave a Technical Report on“Towards Performance Portability for Heterogeneous Systems”

撰稿: 摄影: 发布时间:2014年11月17日
       Christophe Dubach 从EPFL(瑞士)的计算机科学院获得理科硕士学位,从爱丁堡大学(英国)信息学院获得博士学位。现任爱丁堡大学计算机系统体系结构研究所讲师(助理研究员)。在2010年,他以客座研究员身份访问了IBM沃森研究中心(美国),并参与LiquidMetal工程。他目前的研究兴趣包括异构系统的高级语言编程模型、计算机体系结构和编译优化的联合设计技术、自适应微处理器以及机器学习在这些方面的应用。
       受到中科院计算所国家体系结构重点实验室的研究员陈云霁的邀请,Christophe Dubach于2014年11月18日下午14:00,在计算所446会议室作了关于“异构系统性能可移植性”的学术报告。
       在报告中,他展示了可以提供高级编程模型、代码可移植性及高性能的新方法。该方法是基于一些简单规则的重写和算法模式的合成。这就使得可以对高级代码进行系统性的变换和优化转化成低级的硬件专用的表达,从而可以得到性能更高的代码。他同时展示了OpenCL的编程模型是如何映射到低级模式和如何自动产生高性能的OpenCL代码,该代码与针对多核CPU和GPU而高度细致调试实现的代码在性能方面不分伯仲。
       报告结束后,与会者积极参与相关问题的讨论。感谢Christophe Dubach博士给我们带来的精彩报告。
      
      Christophe Dubach received his Ph.D in Informatics from the University of Edinburgh in 2009 and holds a M.Sc. degree in Computer Science from EPFL (Switzerland). He is a Lecturer (Assistant Professor) in the Institute for Computing Systems Architecture at the University of Edinburgh (UK). In 2010 he spent one year as a visiting researcher at the IBM Watson Research Center (USA) working on the LiquidMetal project. His current research interests includes high-level programming models for heterogeneous systems, co-design of both computer architecture and optimising compiler technology, adaptive microprocessor, and the application of machine learning in these areas.
Invited by Yunji Chen, Professor of Key Laboratory of Computer System and Architecture ,Institute of Computing Technology, Chinese Academy of Sciences, Dr. Christophe Dubach gave a report on “Towards Performance Portability for Heterogeneous Systems” on Nov 18, 2014. The report calculated on 14:00PM at Room 446.
    In this talk he presents a novel approach that offers high-level programming, code portability and high-performance. It is based on algorithmic pattern composition coupled with a powerful, yet simple, set of rewrite rules. This enables systematic transformation and optimization of a high-level program into a low-level hardware specific representation which leads to high performance code. He shows how a subset of the OpenCL programming model can be mapped to low-level patterns and how to automatically generate high performance OpenCL code on par with highly tuned implementations for multicore CPUs and GPUs.
After the talk, attendees carried out lively discussion on related issues. Thanks to Dr. Christophe Dubach for bringing us this
excellent talk.
     

附件下载: