CompilerGym: Robust, Performant Compiler Optimization Environments for AI Research

Interest in applying Artificial Intelligence (AI) techniques to compiler optimizations is increasing rapidly, but compiler research has a high entry barrier. Unlike in other domains, compiler and AI researchers do not have access to the datasets and frameworks that enable fast iteration and developm...

Celý popis

Uložené v:
Podrobná bibliografia
Vydané v:2022 IEEE/ACM International Symposium on Code Generation and Optimization (CGO) s. 92 - 105
Hlavní autori: Cummins, Chris, Wasti, Bram, Guo, Jiadong, Cui, Brandon, Ansel, Jason, Gomez, Sahir, Jain, Somya, Liu, Jia, Teytaud, Olivier, Steiner, Benoit, Tian, Yuandong, Leather, Hugh
Médium: Konferenčný príspevok..
Jazyk:English
Vydavateľské údaje: IEEE 02.04.2022
Predmet:
Artificial intelligence
Computational efficiency
Computer bugs
Fault tolerance
Fault tolerant systems
Production
Reproducibility of results
On-line prístup:Získať plný text
Tagy: Pridať tag
Žiadne tagy, Buďte prvý, kto otaguje tento záznam!
Abstract Interest in applying Artificial Intelligence (AI) techniques to compiler optimizations is increasing rapidly, but compiler research has a high entry barrier. Unlike in other domains, compiler and AI researchers do not have access to the datasets and frameworks that enable fast iteration and development of ideas, and getting started requires a significant engineering investment. What is needed is an easy, reusable experimental infrastructure for real world compiler optimization tasks that can serve as a common benchmark for comparing techniques, and as a platform to accelerate progress in the field.We introduce CompilerGym, a set of environments for real world compiler optimization tasks, and a toolkit for exposing new optimization tasks to compiler researchers. CompilerGym enables anyone to experiment on production compiler optimization problems through an easy-to-use package, regardless of their experience with compilers. We build upon the popular OpenAI Gym interface enabling researchers to interact with compilers using Python and a familiar API.We describe the CompilerGym architecture and implementation, characterize the optimization spaces and computational efficiencies of three included compiler environments, and provide extensive empirical evaluations. Compared to prior works, CompilerGym offers larger datasets and optimization spaces, is 27× more computationally efficient, is fault-tolerant, and capable of detecting reproducibility bugs in the underlying compilers.In making it easy for anyone to experiment with compilers - irrespective of their background - we aim to accelerate progress in the AI and compiler research domains.
AbstractList Interest in applying Artificial Intelligence (AI) techniques to compiler optimizations is increasing rapidly, but compiler research has a high entry barrier. Unlike in other domains, compiler and AI researchers do not have access to the datasets and frameworks that enable fast iteration and development of ideas, and getting started requires a significant engineering investment. What is needed is an easy, reusable experimental infrastructure for real world compiler optimization tasks that can serve as a common benchmark for comparing techniques, and as a platform to accelerate progress in the field.We introduce CompilerGym, a set of environments for real world compiler optimization tasks, and a toolkit for exposing new optimization tasks to compiler researchers. CompilerGym enables anyone to experiment on production compiler optimization problems through an easy-to-use package, regardless of their experience with compilers. We build upon the popular OpenAI Gym interface enabling researchers to interact with compilers using Python and a familiar API.We describe the CompilerGym architecture and implementation, characterize the optimization spaces and computational efficiencies of three included compiler environments, and provide extensive empirical evaluations. Compared to prior works, CompilerGym offers larger datasets and optimization spaces, is 27× more computationally efficient, is fault-tolerant, and capable of detecting reproducibility bugs in the underlying compilers.In making it easy for anyone to experiment with compilers - irrespective of their background - we aim to accelerate progress in the AI and compiler research domains.
Author Ansel, Jason
Cui, Brandon
Teytaud, Olivier
Wasti, Bram
Tian, Yuandong
Cummins, Chris
Guo, Jiadong
Gomez, Sahir
Leather, Hugh
Jain, Somya
Liu, Jia
Steiner, Benoit
Author_xml – sequence: 1
  givenname: Chris
  surname: Cummins
  fullname: Cummins, Chris
  email: cummins@fb.com
  organization: Meta,USA
– sequence: 2
  givenname: Bram
  surname: Wasti
  fullname: Wasti, Bram
  organization: Meta,USA
– sequence: 3
  givenname: Jiadong
  surname: Guo
  fullname: Guo, Jiadong
  organization: Meta,USA
– sequence: 4
  givenname: Brandon
  surname: Cui
  fullname: Cui, Brandon
  organization: Meta,USA
– sequence: 5
  givenname: Jason
  surname: Ansel
  fullname: Ansel, Jason
  organization: Meta,USA
– sequence: 6
  givenname: Sahir
  surname: Gomez
  fullname: Gomez, Sahir
  organization: Meta,USA
– sequence: 7
  givenname: Somya
  surname: Jain
  fullname: Jain, Somya
  organization: Meta,USA
– sequence: 8
  givenname: Jia
  surname: Liu
  fullname: Liu, Jia
  organization: Meta,USA
– sequence: 9
  givenname: Olivier
  surname: Teytaud
  fullname: Teytaud, Olivier
  organization: Meta,USA
– sequence: 10
  givenname: Benoit
  surname: Steiner
  fullname: Steiner, Benoit
  organization: Meta,USA
– sequence: 11
  givenname: Yuandong
  surname: Tian
  fullname: Tian, Yuandong
  organization: Meta,USA
– sequence: 12
  givenname: Hugh
  surname: Leather
  fullname: Leather, Hugh
  organization: Meta,USA
BookMark eNo1j99KwzAchSMo6OaeQIQ8gK35n1-8G2Wrg0Gl6PVI2xQDa1rSKMynd-C8OnC-wwdnga7DGBxCj5TklBLzXJSV5IawnBHGcqMFZRKu0IIqJQWRIOAWrebZN0SA4loTeofqYhwmf3SxPA0vuB6brzk94TcX-zEONiT8z3E1JT_4H5v8GPAmfPs4hsGFNOPzFK93uHazs7H9vEc3vT3ObnXJJfrYbt6L12xflbtivc8sA50ypU2neCtaZrmDhnZagISGGApWMWqb9lxq3gATPVOy06BI1zpDJbTQGcKX6OHP651zhyn6wcbT4XKb_wLzwVCM
ContentType Conference Proceeding
DBID 6IE
6IL
CBEJK
RIE
RIL
DOI 10.1109/CGO53902.2022.9741258
DatabaseName IEEE Electronic Library (IEL) Conference Proceedings
IEEE Proceedings Order Plan All Online (POP All Online) 1998-present by volume
IEEE Xplore All Conference Proceedings
IEEE Electronic Library (IEL)
IEEE Proceedings Order Plans (POP All) 1998-Present
DatabaseTitleList
Database_xml – sequence: 1
  dbid: RIE
  name: IEEE Electronic Library (IEL)
  url: https://ieeexplore.ieee.org/
  sourceTypes: Publisher
DeliveryMethod fulltext_linktorsrc
EISBN 1665405848
9781665405843
EndPage 105
ExternalDocumentID 9741258
Genre orig-research
GroupedDBID 6IE
6IL
ACM
ALMA_UNASSIGNED_HOLDINGS
APO
CBEJK
GUFHI
LHSKQ
RIE
RIL
ID FETCH-LOGICAL-a287t-679d63c4c2a3e8b1d74858b0918a621abc8b173b824f265d7860dce9158c8d903
IEDL.DBID RIE
ISICitedReferencesCount 29
ISICitedReferencesURI http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=000827636600008&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
IngestDate Wed Aug 27 02:35:35 EDT 2025
IsPeerReviewed false
IsScholarly true
Language English
LinkModel DirectLink
MergedId FETCHMERGED-LOGICAL-a287t-679d63c4c2a3e8b1d74858b0918a621abc8b173b824f265d7860dce9158c8d903
PageCount 14
ParticipantIDs ieee_primary_9741258
PublicationCentury 2000
PublicationDate 2022-April-2
PublicationDateYYYYMMDD 2022-04-02
PublicationDate_xml – month: 04
  year: 2022
  text: 2022-April-2
  day: 02
PublicationDecade 2020
PublicationTitle 2022 IEEE/ACM International Symposium on Code Generation and Optimization (CGO)
PublicationTitleAbbrev CGO
PublicationYear 2022
Publisher IEEE
Publisher_xml – name: IEEE
SSID ssib048637701
Score 2.4636374
Snippet Interest in applying Artificial Intelligence (AI) techniques to compiler optimizations is increasing rapidly, but compiler research has a high entry barrier....
SourceID ieee
SourceType Publisher
StartPage 92
SubjectTerms Artificial intelligence
Computational efficiency
Computer bugs
Fault tolerance
Fault tolerant systems
Production
Reproducibility of results
Title CompilerGym: Robust, Performant Compiler Optimization Environments for AI Research
URI https://ieeexplore.ieee.org/document/9741258
WOSCitedRecordID wos000827636600008&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
hasFullText 1
inHoldings 1
isFullTextHit
isPrint
link http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlZ1LSwMxEIBDWzx4UmnFNzl4bNpsdvPyJqVWQdpSFHoreUzBQ1tpt4L_3mS7WxW8eAt5EJg8JpOZL0Ho1ijlrZxLYsIGSTJjgVgOlnAw1HMRfb0FKPwsh0M1nepxDbX3LAwAFMFn0InJwpfvV24br8q64ewb9LGqo7qUcsdqVXMnUyKVkiYlpJNQ3e0NRjxY9JG2YqxTtv31iUqhQx6O_tf7MWp9w3h4vFczJ6gGyyaaxHUcFvR68Lm4w5OV3W7yNh5XFECOq3I8CnvCooQtcf8H1oZDVXz_hKvYuxZ6fei_9B5J-T0CMcHMyYmQ2ovUZY6ZFJRNvMwUVzYcAJQRLDHWhUyZWsWyORPcSyWod6ATrpzymqanqLFcLeEMYcGonYMSxsqg0ZnRwINtBppqT70Q2TlqRnnM3ncvYMxKUVz8nX2JDqPIi_gWdoUa-XoL1-jAfeRvm_VNMWxf-geZXQ
linkProvider IEEE
linkToHtml http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlZ1LTwIxEIAniCZ6UgPGtz14ZKHb7dObMbwiAiGYcCPttiQeAAOLif_e7rKLmnjx1vSRJtPHdDrztQD3WkprxEwE2m-QAdXGBYY5EzCnsWU89fVmoHBP9PtyMlHDEtR2LIxzLgs-c_U0mfny7TLepFdlDX_29fpY7sE-o5SEW1qrmD1U8kgIHOaYTohV46k9YN6mT3krQup561_fqGRapHX8v_5PoPqN46HhTtGcQsktKjBKV7Jf0qv25_wBjZZms05qaFhwAAkqytHA7wrzHLdEzR9gG_JV0WMXFdF3VXhtNcdPnSD_ICHQ3tBJAi6U5VFMY6IjJ01oBZVMGn8EkJqTUJvYZ4rISEJnhDMrJMc2dipkMpZW4egMyovlwp0D4gSbmZNcG-F1OtHKMW-dOYWVxZZzegGVVB7T9-0bGNNcFJd_Z9_BYWf80pv2uv3nKzhKxZ9Fu5BrKCerjbuBg_gjeVuvbrMh_AIly5yk
openUrl ctx_ver=Z39.88-2004&ctx_enc=info%3Aofi%2Fenc%3AUTF-8&rfr_id=info%3Asid%2Fsummon.serialssolutions.com&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Abook&rft.genre=proceeding&rft.title=2022+IEEE%2FACM+International+Symposium+on+Code+Generation+and+Optimization+%28CGO%29&rft.atitle=CompilerGym%3A+Robust%2C+Performant+Compiler+Optimization+Environments+for+AI+Research&rft.au=Cummins%2C+Chris&rft.au=Wasti%2C+Bram&rft.au=Guo%2C+Jiadong&rft.au=Cui%2C+Brandon&rft.date=2022-04-02&rft.pub=IEEE&rft.spage=92&rft.epage=105&rft_id=info:doi/10.1109%2FCGO53902.2022.9741258&rft.externalDocID=9741258