TU Berlin

Architektur eingebetteter SystemeMaier, Daniel

AES Logo

Inhalt

zur Navigation

Daniel Maier

Kontaktdaten

Kontaktdaten
Raum:
E-N 638
Tel.:
+49 (0)30 314-25390
E-Mail

Sprechstunde:
nach Vereinbarung

Anschrift:
Sekretariat EN 12
Einsteinufer 17
D-10587 Berlin

Abschlussarbeiten

  • currently no open projects available

Forschung

  • Approximate computing
  • Software- und Compiler-Optimierungen

Awards

  • Outstanding Paper Award Runner Up auf der International Conference on High Performance Computing & Simulation (HPCS 2019) für das Paper "Approximating Memory-bound Applications on Mobile GPUs"
  • Best Paper Award auf der Eighth EAI International Conference on Simulation Tools and Techniques (SIMUTools 2015) für das Paper "Deterministic Models of the Physical Layer through Signal Simulation"

Lehre

Lehrveranstaltungen
Rechnerorganisation
WS 16/17
AES Bachelor Seminar
SS 17
Rechnerorganisation
WS 17/18
AES Bachelor Seminar
SS 18

Publikationen

ALONA: Automatic Loop Nest Approximation with Reconstruction and Space Pruning
Zitatschlüssel 10.1007/978-3-030-85665-6_1
Autor Daniel Maier and Biagio Cosenza and Ben Juurlink
Buchtitel Euro-Par 2021: Parallel Processing
Seiten 3–18
Jahr 2021
ISBN 978-3-030-85665-6
Adresse Cham
Herausgeber Leonel Sousa and Nuno Roma and Pedro Tomás
Verlag Springer International Publishing
Zusammenfassung Approximate computing comprises a large variety of techniques that trade the accuracy of an application's output for other metrics such as computing time or energy cost. Many existing approximation techniques focus on loops such as loop perforation, which skips iterations for faster, approximated computation. This paper introduces ALONA, a novel approach for automatic loop nest approximation based on polyhedral compilation. ALONA's compilation framework applies a sequence of loop approximation transformations, generalizes state-of-the-art perforation techniques, and introduces new multi-dimensional approximation schemes. The framework includes a reconstruction technique that significantly improves the accuracy of the approximations and a transformation space pruning method based on Barvinok's counting that removes inaccurate approximations. Evaluated on a collection of more than twenty applications from PolyBench/C, ALONA discovers new approximations that are better than state-of-the-art techniques in both approximation accuracy and performance.
Download Bibtex Eintrag

Navigation

Direktzugang

Schnellnavigation zur Seite über Nummerneingabe