RScaLAPACK: High-Performance Parallel Statistical Computing with R and ScaLAPACK

Srikanth Yoginath; Nagiza F. Samatova; David Bauer; Guruprasad Kora; George Fann; Al Geist

RScaLAPACK: High-Performance Parallel Statistical Computing with R and ScaLAPACK

Srikanth Yoginath, Nagiza F. Samatova, David Bauer, Guruprasad Kora, George Fann, Al Geist

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

7 Scopus citations

Abstract

With the growing popularity of parallel computation, researchers are looking for various means to reduce the problem solving time by performing the computations in parallel. While benefiting from parallel computation, they do not want to deal with the parallel programming complexities. In this paper, through RScaLAPACK we demonstrate a means that enables the user to carryout parallel computation without dealing with the intricacies of parallel programming. RScaLAPACK is built on top of two Open source software packages: R and ScaLAPACK. The former provides an environment and a language for statistical computing and graphics. The latter is a library of high-performance parallel linear algebra routines using PVM and/or MPI. Through RScaLAPACK the user can setup the parallel environment, distribute data and carry out the required parallel computation using a single R function call. While the interface maintains look and feel of the R system, RScaLAPACK allows carrying out analyses with a performance that scales well with both the problem size and the number of processors. RScaLAPACK is made available as an Open Source package and can be found at http://www.aspect-sdm.org/Parallel-R or on R’s CRAN web site http://www.r-project.org. In this paper we discuss the design, working and performance characteristics of RScaLAPACK.

Original language	English
Title of host publication	18th ISCA International Conference on Parallel and Distributed Computing Systems 2005, PDCS 2005
Publisher	International Society for Computers and Their Applications (ISCA)
Pages	61-67
Number of pages	7
ISBN (Electronic)	9781604234565
State	Published - 2005
Event	18th International Conference on Parallel and Distributed Computing Systems, PDCS 2005 - Las Vegas, United States Duration: Sep 12 2005 → Sep 14 2005

Publication series

Name	18th ISCA International Conference on Parallel and Distributed Computing Systems 2005, PDCS 2005

Conference

Conference	18th International Conference on Parallel and Distributed Computing Systems, PDCS 2005
Country/Territory	United States
City	Las Vegas
Period	09/12/05 → 09/14/05

Keywords

ScaLAPACK
parallel R
parallel statistical computing

Cite this

Yoginath, S., Samatova, N. F., Bauer, D., Kora, G., Fann, G., & Geist, A. (2005). RScaLAPACK: High-Performance Parallel Statistical Computing with R and ScaLAPACK. In 18th ISCA International Conference on Parallel and Distributed Computing Systems 2005, PDCS 2005 (pp. 61-67). (18th ISCA International Conference on Parallel and Distributed Computing Systems 2005, PDCS 2005). International Society for Computers and Their Applications (ISCA).

Yoginath, Srikanth ; Samatova, Nagiza F. ; Bauer, David et al. / RScaLAPACK : High-Performance Parallel Statistical Computing with R and ScaLAPACK. 18th ISCA International Conference on Parallel and Distributed Computing Systems 2005, PDCS 2005. International Society for Computers and Their Applications (ISCA), 2005. pp. 61-67 (18th ISCA International Conference on Parallel and Distributed Computing Systems 2005, PDCS 2005).

@inproceedings{e90100d2b3e84cfcb05121e06798fcf2,

title = "RScaLAPACK: High-Performance Parallel Statistical Computing with R and ScaLAPACK",

abstract = "With the growing popularity of parallel computation, researchers are looking for various means to reduce the problem solving time by performing the computations in parallel. While benefiting from parallel computation, they do not want to deal with the parallel programming complexities. In this paper, through RScaLAPACK we demonstrate a means that enables the user to carryout parallel computation without dealing with the intricacies of parallel programming. RScaLAPACK is built on top of two Open source software packages: R and ScaLAPACK. The former provides an environment and a language for statistical computing and graphics. The latter is a library of high-performance parallel linear algebra routines using PVM and/or MPI. Through RScaLAPACK the user can setup the parallel environment, distribute data and carry out the required parallel computation using a single R function call. While the interface maintains look and feel of the R system, RScaLAPACK allows carrying out analyses with a performance that scales well with both the problem size and the number of processors. RScaLAPACK is made available as an Open Source package and can be found at http://www.aspect-sdm.org/Parallel-R or on R{\textquoteright}s CRAN web site http://www.r-project.org. In this paper we discuss the design, working and performance characteristics of RScaLAPACK.",

keywords = "ScaLAPACK, parallel R, parallel statistical computing",

author = "Srikanth Yoginath and Samatova, {Nagiza F.} and David Bauer and Guruprasad Kora and George Fann and Al Geist",

note = "Publisher Copyright: {\textcopyright} 2005 18th ISCA International Conference on Parallel and Distributed Computing Systems 2005, PDCS 2005. All rights reserved.; 18th International Conference on Parallel and Distributed Computing Systems, PDCS 2005 ; Conference date: 12-09-2005 Through 14-09-2005",

year = "2005",

language = "English",

series = "18th ISCA International Conference on Parallel and Distributed Computing Systems 2005, PDCS 2005",

publisher = "International Society for Computers and Their Applications (ISCA)",

pages = "61--67",

booktitle = "18th ISCA International Conference on Parallel and Distributed Computing Systems 2005, PDCS 2005",

}

Yoginath, S, Samatova, NF, Bauer, D, Kora, G, Fann, G & Geist, A 2005, RScaLAPACK: High-Performance Parallel Statistical Computing with R and ScaLAPACK. in 18th ISCA International Conference on Parallel and Distributed Computing Systems 2005, PDCS 2005. 18th ISCA International Conference on Parallel and Distributed Computing Systems 2005, PDCS 2005, International Society for Computers and Their Applications (ISCA), pp. 61-67, 18th International Conference on Parallel and Distributed Computing Systems, PDCS 2005, Las Vegas, United States, 09/12/05.

RScaLAPACK: High-Performance Parallel Statistical Computing with R and ScaLAPACK. / Yoginath, Srikanth; Samatova, Nagiza F.; Bauer, David et al.
18th ISCA International Conference on Parallel and Distributed Computing Systems 2005, PDCS 2005. International Society for Computers and Their Applications (ISCA), 2005. p. 61-67 (18th ISCA International Conference on Parallel and Distributed Computing Systems 2005, PDCS 2005).

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

TY - GEN

T1 - RScaLAPACK

T2 - 18th International Conference on Parallel and Distributed Computing Systems, PDCS 2005

AU - Yoginath, Srikanth

AU - Samatova, Nagiza F.

AU - Bauer, David

AU - Kora, Guruprasad

AU - Fann, George

AU - Geist, Al

PY - 2005

Y1 - 2005

N2 - With the growing popularity of parallel computation, researchers are looking for various means to reduce the problem solving time by performing the computations in parallel. While benefiting from parallel computation, they do not want to deal with the parallel programming complexities. In this paper, through RScaLAPACK we demonstrate a means that enables the user to carryout parallel computation without dealing with the intricacies of parallel programming. RScaLAPACK is built on top of two Open source software packages: R and ScaLAPACK. The former provides an environment and a language for statistical computing and graphics. The latter is a library of high-performance parallel linear algebra routines using PVM and/or MPI. Through RScaLAPACK the user can setup the parallel environment, distribute data and carry out the required parallel computation using a single R function call. While the interface maintains look and feel of the R system, RScaLAPACK allows carrying out analyses with a performance that scales well with both the problem size and the number of processors. RScaLAPACK is made available as an Open Source package and can be found at http://www.aspect-sdm.org/Parallel-R or on R’s CRAN web site http://www.r-project.org. In this paper we discuss the design, working and performance characteristics of RScaLAPACK.

AB - With the growing popularity of parallel computation, researchers are looking for various means to reduce the problem solving time by performing the computations in parallel. While benefiting from parallel computation, they do not want to deal with the parallel programming complexities. In this paper, through RScaLAPACK we demonstrate a means that enables the user to carryout parallel computation without dealing with the intricacies of parallel programming. RScaLAPACK is built on top of two Open source software packages: R and ScaLAPACK. The former provides an environment and a language for statistical computing and graphics. The latter is a library of high-performance parallel linear algebra routines using PVM and/or MPI. Through RScaLAPACK the user can setup the parallel environment, distribute data and carry out the required parallel computation using a single R function call. While the interface maintains look and feel of the R system, RScaLAPACK allows carrying out analyses with a performance that scales well with both the problem size and the number of processors. RScaLAPACK is made available as an Open Source package and can be found at http://www.aspect-sdm.org/Parallel-R or on R’s CRAN web site http://www.r-project.org. In this paper we discuss the design, working and performance characteristics of RScaLAPACK.

KW - ScaLAPACK

KW - parallel R

KW - parallel statistical computing

UR - http://www.scopus.com/inward/record.url?scp=34548784300&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:34548784300

T3 - 18th ISCA International Conference on Parallel and Distributed Computing Systems 2005, PDCS 2005

SP - 61

EP - 67

BT - 18th ISCA International Conference on Parallel and Distributed Computing Systems 2005, PDCS 2005

PB - International Society for Computers and Their Applications (ISCA)

Y2 - 12 September 2005 through 14 September 2005

ER -

Yoginath S, Samatova NF, Bauer D, Kora G, Fann G, Geist A. RScaLAPACK: High-Performance Parallel Statistical Computing with R and ScaLAPACK. In 18th ISCA International Conference on Parallel and Distributed Computing Systems 2005, PDCS 2005. International Society for Computers and Their Applications (ISCA). 2005. p. 61-67. (18th ISCA International Conference on Parallel and Distributed Computing Systems 2005, PDCS 2005).

RScaLAPACK: High-Performance Parallel Statistical Computing with R and ScaLAPACK

Abstract

Publication series

Conference

Keywords

Other files and links

Fingerprint

Cite this