Academic Journal

RAAS: Runtime Adaptive Approximation System.

Bibliographic Details
Title: RAAS: Runtime Adaptive Approximation System.
Authors: Reis, Lucas, Rigo, Sandro, Wanner, Lucas
Source: Concurrency & Computation: Practice & Experience; Jun2026, Vol. 38 Issue 12, p1-20, 20p
Subject Terms: Compilers (Computer programs), Approximation algorithms, Mathematical optimization
Abstract: Software‐level approximation techniques, such as loop perforation and functional replacement, can improve energy and performance in error‐tolerant applications, but typically require extensive programmer intervention even when compiler support is available. Design‐time and compile‐time approximation approaches also make static decisions, limiting the ability to adapt approximation levels dynamically in response to runtime conditions or inputs. We present RAAS, a Runtime Adaptive Approximation System that identifies and applies approximation opportunities dynamically within a Just‐In‐Time compiler. The programmer marks approximation regions using simple annotations, and provides either built‐in or custom quality metrics along with acceptable error bounds. RAAS explores the approximation space at runtime, evaluating candidate transformations for performance and accuracy, and converges to configurations that maximize speedup while respecting the allowed error threshold. Evaluation on application kernels from domains such as simulation and image processing shows speedups of up to 4.6×$$ 4.6\times $$ with quality degradation below 10%, outperforming static compilation approaches even when accounting for runtime adaptation overhead. [ABSTRACT FROM AUTHOR]
Copyright of Concurrency & Computation: Practice & Experience is the property of Wiley-Blackwell and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)
Database: Complementary Index
FullText Text:
  Availability: 0
CustomLinks:
  – Url: https://resolver.ebsco.com/c/fiv2js/result?sid=EBSCO:edb&genre=article&issn=15320626&ISBN=&volume=38&issue=12&date=20260615&spage=1&pages=1-20&title=Concurrency & Computation: Practice & Experience&atitle=RAAS%3A%20Runtime%20Adaptive%20Approximation%20System.&aulast=Reis%2C%20Lucas&id=DOI:10.1002/cpe.70797
    Name: Full Text Finder (for New FTF UI) (ns324271)
    Category: fullText
    Text: Full Text Finder
    MouseOverText: Full Text Finder
Header DbId: edb
DbLabel: Complementary Index
An: 194723254
RelevancyScore: 1082
AccessLevel: 6
PubType: Academic Journal
PubTypeId: academicJournal
PreciseRelevancyScore: 1082.41931152344
IllustrationInfo
Items – Name: Title
  Label: Title
  Group: Ti
  Data: RAAS: Runtime Adaptive Approximation System.
– Name: Author
  Label: Authors
  Group: Au
  Data: <searchLink fieldCode="AR" term="%22Reis%2C+Lucas%22">Reis, Lucas</searchLink><br /><searchLink fieldCode="AR" term="%22Rigo%2C+Sandro%22">Rigo, Sandro</searchLink><br /><searchLink fieldCode="AR" term="%22Wanner%2C+Lucas%22">Wanner, Lucas</searchLink>
– Name: TitleSource
  Label: Source
  Group: Src
  Data: Concurrency & Computation: Practice & Experience; Jun2026, Vol. 38 Issue 12, p1-20, 20p
– Name: Subject
  Label: Subject Terms
  Group: Su
  Data: <searchLink fieldCode="DE" term="%22Compilers+%28Computer+programs%29%22">Compilers (Computer programs)</searchLink><br /><searchLink fieldCode="DE" term="%22Approximation+algorithms%22">Approximation algorithms</searchLink><br /><searchLink fieldCode="DE" term="%22Mathematical+optimization%22">Mathematical optimization</searchLink>
– Name: Abstract
  Label: Abstract
  Group: Ab
  Data: Software‐level approximation techniques, such as loop perforation and functional replacement, can improve energy and performance in error‐tolerant applications, but typically require extensive programmer intervention even when compiler support is available. Design‐time and compile‐time approximation approaches also make static decisions, limiting the ability to adapt approximation levels dynamically in response to runtime conditions or inputs. We present RAAS, a Runtime Adaptive Approximation System that identifies and applies approximation opportunities dynamically within a Just‐In‐Time compiler. The programmer marks approximation regions using simple annotations, and provides either built‐in or custom quality metrics along with acceptable error bounds. RAAS explores the approximation space at runtime, evaluating candidate transformations for performance and accuracy, and converges to configurations that maximize speedup while respecting the allowed error threshold. Evaluation on application kernels from domains such as simulation and image processing shows speedups of up to 4.6×$$ 4.6\times $$ with quality degradation below 10%, outperforming static compilation approaches even when accounting for runtime adaptation overhead. [ABSTRACT FROM AUTHOR]
– Name: Abstract
  Label:
  Group: Ab
  Data: <i>Copyright of Concurrency & Computation: Practice & Experience is the property of Wiley-Blackwell and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract.</i> (Copyright applies to all Abstracts.)
PLink https://search.ebscohost.com/login.aspx?direct=true&site=eds-live&db=edb&AN=194723254
RecordInfo BibRecord:
  BibEntity:
    Identifiers:
      – Type: doi
        Value: 10.1002/cpe.70797
    Languages:
      – Code: eng
        Text: English
    PhysicalDescription:
      Pagination:
        PageCount: 20
        StartPage: 1
    Subjects:
      – SubjectFull: Compilers (Computer programs)
        Type: general
      – SubjectFull: Approximation algorithms
        Type: general
      – SubjectFull: Mathematical optimization
        Type: general
    Titles:
      – TitleFull: RAAS: Runtime Adaptive Approximation System.
        Type: main
  BibRelationships:
    HasContributorRelationships:
      – PersonEntity:
          Name:
            NameFull: Reis, Lucas
      – PersonEntity:
          Name:
            NameFull: Rigo, Sandro
      – PersonEntity:
          Name:
            NameFull: Wanner, Lucas
    IsPartOfRelationships:
      – BibEntity:
          Dates:
            – D: 15
              M: 06
              Text: Jun2026
              Type: published
              Y: 2026
          Identifiers:
            – Type: issn-print
              Value: 15320626
          Numbering:
            – Type: volume
              Value: 38
            – Type: issue
              Value: 12
          Titles:
            – TitleFull: Concurrency & Computation: Practice & Experience
              Type: main
ResultId 1