Run-time Data Analysis to Drive Compiler Optimizations

Sebastian Kloibhofer

doi:10.1145/3484271.3484974

Run-time Data Analysis to Drive Compiler Optimizations

Sebastian Kloibhofer

Institute for Systems Software

Research output: Chapter in Book/Report/Conference proceeding › Conference proceedings › peer-review

Abstract

Dynamic compilers collect a variety of information to optimize programs and achieve peak performance. Nevertheless, particularly in data-heavy applications, analysis of the processed data - data structures, metrics, relations - could enable additional optimizations in terms of access patterns and data locality. Query planning in database systems is one source of inspiration, but due to the required overhead to collect such information, it is infeasible in dynamic compilers. With this project, we propose integrating data analysis into a dynamic runtime to speed up big data applications. The goal is to use the detailed run-time information for speculative compiler optimizations based on the shape and complexion of the data to improve performance.

Original language	English
Title of host publication	SPLASH Companion 2021: Companion Proceedings of the 2021 ACM SIGPLAN International Conference on Systems, Programming, Languages, and Applications: Software for Humanity
Editors	Hridesh Rajan
Publisher	ACM Digital
Pages	9-12
Number of pages	4
ISBN (Electronic)	9781450390880
DOIs	https://doi.org/10.1145/3484271.3484974
Publication status	Published - Oct 2021

Fields of science

102 Computer Sciences
102009 Computer simulation
102011 Formal languages
102013 Human-computer interaction
102022 Software development
102024 Usability research
102029 Practical computer science

JKU Focus areas

Digital Transformation

Access to Document

10.1145/3484271.3484974

Cite this

@inproceedings{2c28d2ab78c945baae60a9858ecf8ea8,

title = "Run-time Data Analysis to Drive Compiler Optimizations",

abstract = "Dynamic compilers collect a variety of information to optimize programs and achieve peak performance. Nevertheless, particularly in data-heavy applications, analysis of the processed data - data structures, metrics, relations - could enable additional optimizations in terms of access patterns and data locality. Query planning in database systems is one source of inspiration, but due to the required overhead to collect such information, it is infeasible in dynamic compilers. With this project, we propose integrating data analysis into a dynamic runtime to speed up big data applications. The goal is to use the detailed run-time information for speculative compiler optimizations based on the shape and complexion of the data to improve performance.",

author = "Sebastian Kloibhofer",

year = "2021",

month = oct,

doi = "10.1145/3484271.3484974",

language = "English",

pages = "9--12",

editor = "Hridesh Rajan",

booktitle = "SPLASH Companion 2021: Companion Proceedings of the 2021 ACM SIGPLAN International Conference on Systems, Programming, Languages, and Applications: Software for Humanity",

publisher = "ACM Digital",

}

Run-time Data Analysis to Drive Compiler Optimizations. / Kloibhofer, Sebastian.
SPLASH Companion 2021: Companion Proceedings of the 2021 ACM SIGPLAN International Conference on Systems, Programming, Languages, and Applications: Software for Humanity. ed. / Hridesh Rajan. ACM Digital, 2021. p. 9-12.

Research output: Chapter in Book/Report/Conference proceeding › Conference proceedings › peer-review

TY - GEN

T1 - Run-time Data Analysis to Drive Compiler Optimizations

AU - Kloibhofer, Sebastian

PY - 2021/10

Y1 - 2021/10

N2 - Dynamic compilers collect a variety of information to optimize programs and achieve peak performance. Nevertheless, particularly in data-heavy applications, analysis of the processed data - data structures, metrics, relations - could enable additional optimizations in terms of access patterns and data locality. Query planning in database systems is one source of inspiration, but due to the required overhead to collect such information, it is infeasible in dynamic compilers. With this project, we propose integrating data analysis into a dynamic runtime to speed up big data applications. The goal is to use the detailed run-time information for speculative compiler optimizations based on the shape and complexion of the data to improve performance.

AB - Dynamic compilers collect a variety of information to optimize programs and achieve peak performance. Nevertheless, particularly in data-heavy applications, analysis of the processed data - data structures, metrics, relations - could enable additional optimizations in terms of access patterns and data locality. Query planning in database systems is one source of inspiration, but due to the required overhead to collect such information, it is infeasible in dynamic compilers. With this project, we propose integrating data analysis into a dynamic runtime to speed up big data applications. The goal is to use the detailed run-time information for speculative compiler optimizations based on the shape and complexion of the data to improve performance.

UR - https://www.scopus.com/pages/publications/85121454616

U2 - 10.1145/3484271.3484974

DO - 10.1145/3484271.3484974

M3 - Conference proceedings

SP - 9

EP - 12

BT - SPLASH Companion 2021: Companion Proceedings of the 2021 ACM SIGPLAN International Conference on Systems, Programming, Languages, and Applications: Software for Humanity

A2 - Rajan, Hridesh

PB - ACM Digital

ER -

Run-time Data Analysis to Drive Compiler Optimizations

Abstract

Fields of science

JKU Focus areas

Access to Document

Other files and links

Java VM Compiler Performance (Oracle)

Cite this

Run-time Data Analysis to Drive Compiler Optimizations

Abstract

Fields of science

JKU Focus areas

Access to Document

Other files and links

Projects

Java VM Compiler Performance (Oracle)

Cite this