GPU accelerated nonlinear optimization in radio interferometric calibration

Sarod Yatawatta; Sanaz Kazemi; Saleem Zaroubi

doi:10.1109/InPar.2012.6339610

GPU accelerated nonlinear optimization in radio interferometric calibration

Sarod Yatawatta, Sanaz Kazemi, Saleem Zaroubi

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

Abstract

We present the GPU based acceleration of two well known nonlinear optimization routines: Levenberg-Marquardt (LM) and Limited Memory Broyden-Fletcher-Goldfarb-Shanno (LBFGS) in radio interferometric calibration. Radio interferometric calibration is a heavily compute intensive operation where the same nonlinear optimization problem has to be solved over many time intervals, with different data. We achieve a speedup of about 3 times compared with conventional multi-core CPU based optimization by using GPU accelerated linear algebra routines (CULAtools,CUBLAS). We present details of our GPU accelerated optimization algorithms as well as timing comparisons with non-GPU based multi-core CPU routines.

Original language	English
Title of host publication	2012 Innovative Parallel Computing, InPar 2012
DOIs	https://doi.org/10.1109/InPar.2012.6339610
State	Published - 2012
Externally published	Yes
Event	2012 Innovative Parallel Computing, InPar 2012 - San Jose, CA, United States Duration: 13 May 2012 → 14 May 2012

Publication series

Name	2012 Innovative Parallel Computing, InPar 2012

Conference

Conference	2012 Innovative Parallel Computing, InPar 2012
Country/Territory	United States
City	San Jose, CA
Period	13/05/12 → 14/05/12

Access to Document

10.1109/InPar.2012.6339610

Cite this

@inproceedings{05b722ac526a4b7c8e73cf6abfe62710,

title = "GPU accelerated nonlinear optimization in radio interferometric calibration",

abstract = "We present the GPU based acceleration of two well known nonlinear optimization routines: Levenberg-Marquardt (LM) and Limited Memory Broyden-Fletcher-Goldfarb-Shanno (LBFGS) in radio interferometric calibration. Radio interferometric calibration is a heavily compute intensive operation where the same nonlinear optimization problem has to be solved over many time intervals, with different data. We achieve a speedup of about 3 times compared with conventional multi-core CPU based optimization by using GPU accelerated linear algebra routines (CULAtools,CUBLAS). We present details of our GPU accelerated optimization algorithms as well as timing comparisons with non-GPU based multi-core CPU routines.",

author = "Sarod Yatawatta and Sanaz Kazemi and Saleem Zaroubi",

year = "2012",

doi = "10.1109/InPar.2012.6339610",

language = "אנגלית",

isbn = "9781467326322",

series = "2012 Innovative Parallel Computing, InPar 2012",

booktitle = "2012 Innovative Parallel Computing, InPar 2012",

}

TY - GEN

T1 - GPU accelerated nonlinear optimization in radio interferometric calibration

AU - Yatawatta, Sarod

AU - Kazemi, Sanaz

AU - Zaroubi, Saleem

PY - 2012

Y1 - 2012

N2 - We present the GPU based acceleration of two well known nonlinear optimization routines: Levenberg-Marquardt (LM) and Limited Memory Broyden-Fletcher-Goldfarb-Shanno (LBFGS) in radio interferometric calibration. Radio interferometric calibration is a heavily compute intensive operation where the same nonlinear optimization problem has to be solved over many time intervals, with different data. We achieve a speedup of about 3 times compared with conventional multi-core CPU based optimization by using GPU accelerated linear algebra routines (CULAtools,CUBLAS). We present details of our GPU accelerated optimization algorithms as well as timing comparisons with non-GPU based multi-core CPU routines.

AB - We present the GPU based acceleration of two well known nonlinear optimization routines: Levenberg-Marquardt (LM) and Limited Memory Broyden-Fletcher-Goldfarb-Shanno (LBFGS) in radio interferometric calibration. Radio interferometric calibration is a heavily compute intensive operation where the same nonlinear optimization problem has to be solved over many time intervals, with different data. We achieve a speedup of about 3 times compared with conventional multi-core CPU based optimization by using GPU accelerated linear algebra routines (CULAtools,CUBLAS). We present details of our GPU accelerated optimization algorithms as well as timing comparisons with non-GPU based multi-core CPU routines.

UR - http://www.scopus.com/inward/record.url?scp=84870693491&partnerID=8YFLogxK

U2 - 10.1109/InPar.2012.6339610

DO - 10.1109/InPar.2012.6339610

M3 - ???researchoutput.researchoutputtypes.contributiontobookanthology.conference???

AN - SCOPUS:84870693491

SN - 9781467326322

T3 - 2012 Innovative Parallel Computing, InPar 2012

BT - 2012 Innovative Parallel Computing, InPar 2012

T2 - 2012 Innovative Parallel Computing, InPar 2012

Y2 - 13 May 2012 through 14 May 2012

ER -

GPU accelerated nonlinear optimization in radio interferometric calibration

Abstract

Publication series

Conference

Access to Document

Other files and links

Fingerprint

Cite this