Academic Journal
RAAS: Runtime Adaptive Approximation System.
| 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 |