Transforming loop chains via macro dataflow graphs

Eddie C. Davis, Michelle Mills Strout, Catherine Olschanowsky

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

11 Scopus citations

Abstract

This paper describes an approach to performance optimization using modified macro dataflow graphs, which contain nodes representing the loops and data involved in the stencil computation. The targeted applications include existing scientific applications that contain a series of stencil computations that share data, i.e. loop chains. The performance of stencil applications can be improved by modifying the execution schedules. However, modern architectures are increasingly constrained by the memory subsystem bandwidth. To fully realize the benefits of the schedule changes for improved locality, temporary storage allocation must also be minimized. We present a macro dataflow graph variant that includes dataset nodes, a cost model that quantifies the memory interactions required by a given graph, a set of transformations that can be performed on the graphs such as fusion and tiling, and an approach for generating code to implement the transformed graph. We include a performance comparison with Halide and PolyMage implementations of the benchmark. Our fastest variant outperforms the auto-tuned variants produced by both frameworks.

Original languageEnglish (US)
Title of host publicationCGO 2018 - Proceedings of the 2018 International Symposium on Code Generation and Optimization
PublisherAssociation for Computing Machinery, Inc
Pages265-277
Number of pages13
ISBN (Electronic)9781450356176
DOIs
StatePublished - Feb 24 2018
Externally publishedYes
Event16th International Symposium on Code Generation and Optimization, CGO 2018 - Vienna, Austria
Duration: Feb 24 2018Feb 28 2018

Publication series

NameCGO 2018 - Proceedings of the 2018 International Symposium on Code Generation and Optimization
Volume2018-February

Conference

Conference16th International Symposium on Code Generation and Optimization, CGO 2018
Country/TerritoryAustria
CityVienna
Period2/24/182/28/18

Keywords

  • Dataflow
  • Loop chain
  • Stencil
  • Storage optimizations

ASJC Scopus subject areas

  • Software
  • Computational Theory and Mathematics
  • Applied Mathematics

Fingerprint

Dive into the research topics of 'Transforming loop chains via macro dataflow graphs'. Together they form a unique fingerprint.

Cite this