AI-Assisted Design-Space Analysis of High-Performance Arm Processors

This work quantifies the impact of microarchitectural features in modern high-performance Arm CPUs. To combat a parameter space that is too large to traverse naively, we employ a decision tree regression machine learning model to predict the number of execution cycles with 93.38% accuracy compared t...

Celý popis

Uloženo v:
Podrobná bibliografie
Vydáno v:SC24-W: Workshops of the International Conference for High Performance Computing, Networking, Storage and Analysis s. 1455 - 1467
Hlavní autoři: Moore, Joseph, Deakin, Tom, McIntosh-Smith, Simon
Médium: Konferenční příspěvek
Jazyk:angličtina
Vydáno: IEEE 17.11.2024
Témata:
On-line přístup:Získat plný text
Tagy: Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
Abstract This work quantifies the impact of microarchitectural features in modern high-performance Arm CPUs. To combat a parameter space that is too large to traverse naively, we employ a decision tree regression machine learning model to predict the number of execution cycles with 93.38% accuracy compared to the simulated cycles. We build on previous work by specializing our design to real-world HPC workloads and modernize our approach with updated search spaces, improved simulation frameworks, and over 180,000 simulated data points. We find empirically that vector length typically has the largest impact on HPC code performance at 25.91% of our performance weighting, followed by memory performance across all levels of the memory hierarchy, and the size of the reorder buffer and register files. Our results motivate deeper exploration of these parameters in both hardware design and simulation, as well as advancing the modelling of architectural simulation through the use of machine learning.
AbstractList This work quantifies the impact of microarchitectural features in modern high-performance Arm CPUs. To combat a parameter space that is too large to traverse naively, we employ a decision tree regression machine learning model to predict the number of execution cycles with 93.38% accuracy compared to the simulated cycles. We build on previous work by specializing our design to real-world HPC workloads and modernize our approach with updated search spaces, improved simulation frameworks, and over 180,000 simulated data points. We find empirically that vector length typically has the largest impact on HPC code performance at 25.91% of our performance weighting, followed by memory performance across all levels of the memory hierarchy, and the size of the reorder buffer and register files. Our results motivate deeper exploration of these parameters in both hardware design and simulation, as well as advancing the modelling of architectural simulation through the use of machine learning.
Author McIntosh-Smith, Simon
Moore, Joseph
Deakin, Tom
Author_xml – sequence: 1
  givenname: Joseph
  surname: Moore
  fullname: Moore, Joseph
  email: zi23956@bristol.ac.uk
  organization: University of Bristol,Bristol,United Kingdom
– sequence: 2
  givenname: Tom
  surname: Deakin
  fullname: Deakin, Tom
  email: tom.deakin@bristol.ac.uk
  organization: University of Bristol,Bristol,United Kingdom
– sequence: 3
  givenname: Simon
  surname: McIntosh-Smith
  fullname: McIntosh-Smith, Simon
  email: s.mcintosh-smith@bristol.ac.uk
  organization: University of Bristol,Bristol,United Kingdom
BookMark eNotjN1KwzAYQCMoqLNPoBd9gdQv-Zq_y9KpGwwcTPFypPHLLKztSHazt3firg6cA-eeXY_TSIw9CqiEAPe8ab80yhoqCbKuAITVV6xwxllUgEqpGm9ZkXPfgQZla7Dqjs2bJW_OLh_pu5xT7ncj3xx8oLIZ_f50DuUUy0W_--FrSnFKgx__YhrKdZoC5Tyl_MBuot9nKi6csc_Xl492wVfvb8u2WXEvlT5y1DpQRK9lpEABOhQQXVRGo1fKWuOtCyi8CsZYSy5Gpx0aYTtEE5TDGXv6__ZEtD2kfvDptBVgJRhp8Bf3sEs0
CODEN IEEPAD
ContentType Conference Proceeding
DBID 6IE
6IL
CBEJK
RIE
RIL
DOI 10.1109/SCW63240.2024.00186
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 Xplore
  url: https://ieeexplore.ieee.org/
  sourceTypes: Publisher
DeliveryMethod fulltext_linktorsrc
EISBN 9798350355543
EndPage 1467
ExternalDocumentID 10820727
Genre orig-research
GroupedDBID 6IE
6IL
ACM
ALMA_UNASSIGNED_HOLDINGS
CBEJK
RIE
RIL
ID FETCH-LOGICAL-a256t-366cef3a62fecec0b310f9f5763a55887a89c31a5c7788e9ff9693718b337c593
IEDL.DBID RIE
ISICitedReferencesCount 0
ISICitedReferencesURI http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=001451792300151&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 01:59:34 EDT 2025
IsPeerReviewed false
IsScholarly false
Language English
LinkModel DirectLink
MergedId FETCHMERGED-LOGICAL-a256t-366cef3a62fecec0b310f9f5763a55887a89c31a5c7788e9ff9693718b337c593
PageCount 13
ParticipantIDs ieee_primary_10820727
PublicationCentury 2000
PublicationDate 2024-Nov.-17
PublicationDateYYYYMMDD 2024-11-17
PublicationDate_xml – month: 11
  year: 2024
  text: 2024-Nov.-17
  day: 17
PublicationDecade 2020
PublicationTitle SC24-W: Workshops of the International Conference for High Performance Computing, Networking, Storage and Analysis
PublicationTitleAbbrev SC-W
PublicationYear 2024
Publisher IEEE
Publisher_xml – name: IEEE
SSID ssib060584085
Score 1.8893807
Snippet This work quantifies the impact of microarchitectural features in modern high-performance Arm CPUs. To combat a parameter space that is too large to traverse...
SourceID ieee
SourceType Publisher
StartPage 1455
SubjectTerms aarch64
Adaptation models
Codes
Computational modeling
Data models
Machine learning
Performance gain
Predictive models
Random access memory
Registers
simeng
simulation
SST
vectorization
Vectors
Title AI-Assisted Design-Space Analysis of High-Performance Arm Processors
URI https://ieeexplore.ieee.org/document/10820727
WOSCitedRecordID wos001451792300151&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/eLvHCXMwlV09T8MwED3RioEJEEV8ywOrIYkTOx5RSwUSqioVRLfKcc4SS4P6we_nzk0pCwNblCXKPSd3z773DuBWE-1xVFZInuwic1M7WWqvpbdVTfVBjVmIJq4vZjQqp1M7bsXqUQuDiLH5DO_4Mp7l141f81YZfeGUryjhdqBjjN6ItbaLh4_32K2rdRZKE3s_6b-zGXlCLDBjj-yUBdO_ZqjEFDI8_OfDj6C3E-OJ8U-aOYY9nJ_A4OFZUmQZo1oMYheGnBD9RbF1GRFNENzEIcc7aYAgWEUrDWgWyx68DR9f-0-yHYggHVUmK6m09hiU01lAjz6pqDYLNhBlUK4o6HfhSutV6gpviNmiDcFq9rsrK6WML6w6he68meMZiEAQYZ5XeYkmt1RjGYKm8sHmjhhaqc6hxyGYfW48L2bbt7_44_4lHHCUWaWXmivorhZrvIZ9_7X6WC5uIlLfEUmTCw
linkProvider IEEE
linkToHtml http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlV09T8MwED1BQYIJEEV844HV0NSJE4-opWpFqSq1iG6V45wllqbqB7-fO5NSFga2KEuUe07unn3vHcC9JtpjqayQPNlFxmlhZaadls7kBdUHBTZ9MHHtp4NBNpmYYSVWD1oYRAzNZ_jAl-EsvyjdmrfK6AunfEUJdxf2eHRWJdfaLB8-4GO_rspbKGqYx1Hrne3IG8QDm-ySHbFk-tcUlZBEOkf_fPwx1LdyPDH8STQnsIOzU2g_9STFllEqRDv0YcgREWAUG58RUXrBbRxyuBUHCAJWVOKAcrGsw1vnedzqymokgrRUm6yk0tqhV1Y3PTp0jZyqM288kQZlk4R-GDYzTkU2cSlxWzTeG82Od1muVOoSo86gNitneA7CE0gYx3mcYRobqrJSAid33sSWOFqmLqDOIZjOv10vppu3v_zj_h0cdMev_Wm_N3i5gkOOOGv2ovQaaqvFGm9g332uPpaL24DaF-4vllQ
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=SC24-W%3A+Workshops+of+the+International+Conference+for+High+Performance+Computing%2C+Networking%2C+Storage+and+Analysis&rft.atitle=AI-Assisted+Design-Space+Analysis+of+High-Performance+Arm+Processors&rft.au=Moore%2C+Joseph&rft.au=Deakin%2C+Tom&rft.au=McIntosh-Smith%2C+Simon&rft.date=2024-11-17&rft.pub=IEEE&rft.spage=1455&rft.epage=1467&rft_id=info:doi/10.1109%2FSCW63240.2024.00186&rft.externalDocID=10820727