Ändra sökning
RefereraExporteraLänk till posten
Permanent länk

Direktlänk
Referera
Referensformat
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Annat format
Fler format
Språk
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Annat språk
Fler språk
Utmatningsformat
  • html
  • text
  • asciidoc
  • rtf
Accuracy-Aware Power Management for Many-Core Systems Running Error-Resilient Applications
Visa övriga samt affilieringar
2017 (Engelska)Ingår i: IEEE Transactions on Very Large Scale Integration (vlsi) Systems, ISSN 1063-8210, E-ISSN 1557-9999, Vol. 25, nr 10, s. 2749-2762Artikel i tidskrift (Refereegranskat) Published
Abstract [en]

Power capping techniques based on dynamic voltage and frequency scaling (DVFS) and power gating (PG) are oriented toward power actuation, compromising on performance and energy. Inherent error resilience of emerging application domains, such as Internet-of-Things (IoT) and machine learning, provides opportunities for energy and performance gains. Leveraging accuracy-performance tradeoffs in such applications, we propose approximation (APPX) as another knob for close-looped power management, to complement power knobs with performance and energy gains. We design a power management framework, APPEND+, that can switch between accurate and approximate modes of execution subject to system throughput requirements. APPEND+ considers the sensitivity of the application to error to make disciplined alteration between levels of APPX such that performance is maximized while error is minimized. We implement a power management scheme that uses APPX, DVFS, and PG knobs hierarchically. We evaluated our proposed approach over machine learning and signal processing applications along with two case studies on IoT-early warning score system and fall detection. APPEND+ yields 1.9x higher throughput, improved latency up to five times, better performance per energy, and dark silicon mitigation compared with the state-of-the-art power management techniques over a set of applications ranging from high to no error resilience.

Ort, förlag, år, upplaga, sidor
IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC , 2017. Vol. 25, nr 10, s. 2749-2762
Nyckelord [en]
Approximate computing, dark silicon, Internet-of-Things (IoT), power management, runtime mapping
Nationell ämneskategori
Datavetenskap (datalogi)
Identifikatorer
URN: urn:nbn:se:kth:diva-217448DOI: 10.1109/TVLSI.2017.2694388ISI: 000413751500007Scopus ID: 2-s2.0-85018969687OAI: oai:DiVA.org:kth-217448DiVA, id: diva2:1158032
Anmärkning

QC 20171117

Tillgänglig från: 2017-11-17 Skapad: 2017-11-17 Senast uppdaterad: 2018-01-13Bibliografiskt granskad

Open Access i DiVA

Fulltext saknas i DiVA

Övriga länkar

Förlagets fulltextScopus

Personposter BETA

Tenhunen, Hannu

Sök vidare i DiVA

Av författaren/redaktören
Tenhunen, Hannu
Av organisationen
Integrerade komponenter och kretsar
I samma tidskrift
IEEE Transactions on Very Large Scale Integration (vlsi) Systems
Datavetenskap (datalogi)

Sök vidare utanför DiVA

GoogleGoogle Scholar

doi
urn-nbn

Altmetricpoäng

doi
urn-nbn
Totalt: 209 träffar
RefereraExporteraLänk till posten
Permanent länk

Direktlänk
Referera
Referensformat
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Annat format
Fler format
Språk
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Annat språk
Fler språk
Utmatningsformat
  • html
  • text
  • asciidoc
  • rtf