Understanding the Power of Evolutionary Computation for GPU Code Optimization

Liou, Jhe-Yu; Awan, Muaaz Gul; Hofmeyr, Steven; Forrest, Stephanie; Wu, Carole-Jean

doi:10.1109/iiswc55918.2022.00025

Search citation statements

Order By: Relevance

Paper Sections

Select...

Introduction1

Citation Types

Supporting

Mentioning

Contrasting

Year Published

2023

Publication Types

Select...

Book1

Relationship

Self Cite0

Independent1

Authors

Journals

Cited by 1 publication

(1 citation statement)

References 38 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We applied genetic programming to human written CUDA (or C) source code, whereas Tony Lewis showed GP could be used to evolve nVidia's PTX GPU assembler [27]. More recently Jhe-Yu Liou et al [28,29,30] have applied grammatical evolution (GE) [31] to LLVM IR for CUDA applications running on nVidia parallel hardware and shown further real world examples where GI finds considerable improvement on hand optimised high level GPU code. GE runs were either for two or seven days.…”

Section: Introductionmentioning

confidence: 99%

Genetic Improvement of LLVM Intermediate Representation

Langdon¹,

Al-Subaihin²,

Blot³

et al. 2023

Lecture Notes in Computer Science

View full text Add to dashboard Cite

Evolving LLVM IR is widely applicable, with LLVM Clang offering support for an increasing range of computer hardware and programming languages. Local search mutations are used to hill climb industry C code released to support geographic open standards: Open Location Code (OLC) from Google and Uber's Hexagonal Hierarchical Spatial Index (H3), giving up to two percent speed up on compiler optimised code.

show abstract