Blogs (61) >>
Tue 17 Jul 2018 14:00 - 14:25 at Hanoi - Session 2

SIMD (Single Instruction, Multiple Data) instruction sets are ubiquitous on modern hardware, but rarely used in software projects. A major reason for this is that efficient SIMD code requires data to be laid out in memory in an unconventional manner, forcing developers to explicitly refactor their code and data structures in order to make use of SIMD. In previous work, we proposed SHAPES, an abstract layout specification for enabling memory optimisations for managed, object-oriented languages. In this paper, we explain how, by extending SHAPES with well-known constructs from the literature, which are not specific to SIMD, we can extend SHAPES to compile programs to use SIMD instructions. The resulting language (sketch) seems able to exploit SIMD capabilities without sacrificing ease of development.

Tue 17 Jul
Times are displayed in time zone: Amsterdam, Berlin, Bern, Rome, Stockholm, Vienna change

14:00 - 15:30: Session 2ICOOOLPS at Hanoi
14:00 - 14:25
Research paper
Extending SHAPES for SIMD Architectures - An approach to native support for Struct of Arrays in languages
Alexandros TasosImperial College London, Juliana FrancoMicrosoft Research, Cambridge, Tobias WrigstadUppsala University, Sophia DrossopoulouImperial College London, Susan EisenbachImperial College London
14:30 - 14:55
Research paper
Maintaining Canonical Form After Edge Deletion
Eric FritzUniversity of Wisconsin, Milwaukee
15:00 - 15:25
Research paper
Shared Memory Implementations of Protocol Programming Languages, Data-Race-Free
Micha HergardenOpen University of the Netherlands, The Netherlands, Sung-Shik JongmansOpen University of the Netherlands, The Netherlands