Data layout inference for code vectorisation

Artjoms Sinkarovs, Sven Bodo Scholz

Research output: Chapter in Book/Report/Conference proceedingConference contribution

2 Citations (Scopus)


SIMD instructions of modern CPUs are crucially important for the performance of compute-intensive algorithms. Auto-vectorisation often fails due to an unfortunate choice of data layout by the programmer. This paper proposes a data layout inference for auto-vectorisation which identifies layout transformations that convert SIMD-unfavorable layouts of data structures into favorable ones. We present a type system for layout transformations and we sketch an inference algorithm for it. Finally, we present some initial performance figures for the impact of the inferred layout transformations. They show that non-intuitive layouts that are inferred through our system can have a vast performance impact on compute intensive programs.

Original languageEnglish
Title of host publicationProceedings of the 2013 International Conference on High Performance Computing and Simulation, HPCS 2013
Number of pages8
Publication statusPublished - 26 Nov 2013
Event2013 11th International Conference on High Performance Computing and Simulation - Helsinki, United Kingdom
Duration: 1 Jul 20135 Jul 2013


Conference2013 11th International Conference on High Performance Computing and Simulation
Abbreviated titleHPCS 2013
Country/TerritoryUnited Kingdom

ASJC Scopus subject areas

  • Applied Mathematics
  • Modelling and Simulation


Dive into the research topics of 'Data layout inference for code vectorisation'. Together they form a unique fingerprint.

Cite this