Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Data Flow Transformation for Energy-Efficient Implementation of Givens Rotation--Based QRD

Data Flow Transformation for Energy-Efficient Implementation of Givens Rotation--Based QRD Data Flow Transformation for Energy-Efficient Implementation of Givens Rotation­Based QRD NAMITA SHARMA and PREETI RANJAN PANDA, Indian Institute of Technology Delhi FRANCKY CATTHOOR, Interuniversity Microelectronics Centre and K. U. Leuven MIN LI, Interuniversity Microelectronics Centre PRASHANT AGRAWAL, Interuniversity Microelectronics Centre and K. U. Leuven QR decomposition (QRD), a matrix decomposition algorithm widely used in embedded application domain, can be realized in a large number of valid processing sequences that differ significantly in the number of memory accesses and computations, and hence the overall implementation energy. With modern low-power embedded processors evolving toward register files with wide memory interfaces and vector functional units (FUs), data flow in these algorithms needs to be carefully devised to efficiently utilize the costly wide memory accesses and the vector FUs. In this article, we present an energy-efficient data flow transformation strategy for the Givens rotation­based QRD. Categories and Subject Descriptors: C.3 [Special Purpose and Application-Based Systems]: RealTime and Embedded Systems; D.3.4 [Programming Languages]: Processors--Compilers, optimization; E.1 [Data Structures]: Arrays General Terms: Design Additional Key Words and Phrases: Data flow transformation, energy optimization, matrix decomposition, SIMD architectures ACM Reference Format: Namita Sharma, Preeti Ranjan Panda, Francky Catthoor, Min Li, and Prashant Agrawal. 2016. Data http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png ACM Transactions on Embedded Computing Systems (TECS) Association for Computing Machinery

Data Flow Transformation for Energy-Efficient Implementation of Givens Rotation--Based QRD

Download PDF
23 pages

Loading next page...
 
/lp/association-for-computing-machinery/data-flow-transformation-for-energy-efficient-implementation-of-givens-JS6tO12oYN

References (26)

Publisher
Association for Computing Machinery
Copyright
Copyright © 2016 by ACM Inc.
ISSN
1539-9087
DOI
10.1145/2837025
Publisher site
See Article on Publisher Site

Abstract

Data Flow Transformation for Energy-Efficient Implementation of Givens Rotation­Based QRD NAMITA SHARMA and PREETI RANJAN PANDA, Indian Institute of Technology Delhi FRANCKY CATTHOOR, Interuniversity Microelectronics Centre and K. U. Leuven MIN LI, Interuniversity Microelectronics Centre PRASHANT AGRAWAL, Interuniversity Microelectronics Centre and K. U. Leuven QR decomposition (QRD), a matrix decomposition algorithm widely used in embedded application domain, can be realized in a large number of valid processing sequences that differ significantly in the number of memory accesses and computations, and hence the overall implementation energy. With modern low-power embedded processors evolving toward register files with wide memory interfaces and vector functional units (FUs), data flow in these algorithms needs to be carefully devised to efficiently utilize the costly wide memory accesses and the vector FUs. In this article, we present an energy-efficient data flow transformation strategy for the Givens rotation­based QRD. Categories and Subject Descriptors: C.3 [Special Purpose and Application-Based Systems]: RealTime and Embedded Systems; D.3.4 [Programming Languages]: Processors--Compilers, optimization; E.1 [Data Structures]: Arrays General Terms: Design Additional Key Words and Phrases: Data flow transformation, energy optimization, matrix decomposition, SIMD architectures ACM Reference Format: Namita Sharma, Preeti Ranjan Panda, Francky Catthoor, Min Li, and Prashant Agrawal. 2016. Data

Journal

ACM Transactions on Embedded Computing Systems (TECS) – Association for Computing Machinery

Published: Jan 13, 2016

There are no references for this article.