Programming with Big Data in R: Difference between revisions

Revision as of 14:55, 2 October 2013

Latest revision as of 16:31, 28 February 2024

(68 intermediate revisions by 38 users not shown)

Line 1:

{{multiple issues|

{{notability|date=June 2013}}

{{COI|date=June 2013}}

}}

{{Expert-subject|Computer science|date=June 2013}}

{{Infobox programming language

| name = pbdRbdrp

| logo = [[File:Pbdr.png|200px]]

| paradigm = [[SPMD]] and [[MPMD]]

| yearreleased = {{Start date and = Sep. age|2012|09}}

| designer = Wei-Chen Chen, George Ostrouchov, Pragneshkumar Patel, and Drew Schmidt

| developer = pbdR Core Team

| latest_test_version = Through [[GitHub]] at [httphttps://github.com/RBigData/ RBigData]

| typing = [[dynamic typing|Dynamic]]

| influenced_by = [[R (programming language)|R]], [[C (programming language)|C]], [[Fortran (programming language)|Fortran]], and [[Message Passing Interface|MPI]], and [[ZeroMQ|ØMQ]]

| operating_system = [[Cross-platform]]

| license = [[General Public License]] and [[Mozilla Public License]]

| website = [http://{{URL|www.r-pbd.org/ http://www.r-pbd.org/]}}

}}

'''Programming with Big Data in R''' (pbdR)<ref>{{cite web|author=Ostrouchov, G., Chen, W.-C., Schmidt, D., Patel, P.|title=Programming with Big Data in R|year=2012|url=http://r-pbd.org}}</ref> is a series of [[R (programming language)|R]] packages and an environment for [[statistical computing]] with [[Bigbig Datadata]] by utilizingusing high-performance statistical computation.<ref>{{cite web|authorauthor1=Chen, W.-C. and |author2=Ostrouchov, G.|name-list-style=amp|url=http://thirteen-01.stat.iastate.edu/snoweye/hpsc/|year=2011|title=HPSC -- High Performance Statistical Computing for Data Intensive Research|access-date=2013-06-25|archive-url=https://web.archive.org/web/20130719020318/http://thirteen-01.stat.iastate.edu/snoweye/hpsc/|archive-date=2013-07-19|url-status=dead}}</ref><ref>{{cite web|url=https://learnshareit.com/tutorials-for-r/|title=Basic Tutorials for R to Start Analyzing Data|date=3 November 2022 }}</ref> The pbdR uses the same programming language as R with [[S (programming language)|S3/S4]] classes and methods which is used among [[statistician]]s and [[Data mining|data miners]] for developing [[statistical software]]. The significant difference between pbdR and R codescode is that pbdR mainly focuses on [[distributed memory]] systemsystems, where data are distributed across several processors and analyzed in a [[Batch processing|batch mode]], while communications between processors are based on [[Message Passing Interface|MPI]] whichthat is easily utilizedused in large [[High-performance computing|high-performance computing (HPC)]] systems. R system mainly focuses{{Citation needed|date=July 2013}} on single [[Multi-core processor| multi-core]] machines for data analysis via an interactive mode such as [[Graphical user interface|GUI interface]].

Two main implementations in [[R (programming language)|R]] using [[Message Passing Interface|MPI]] are Rmpi<ref name=rmpi>{{cite journal|author=Yu, H.|title=Rmpi: Parallel Statistical Computing in R|year=2002|url=http://cran.r-project.org/package=Rmpi|journal=R News}}</ref> and pbdMPI of pbdR.

* The pbdR built on pbdMPI uses [[SPMD|SPMD parallelism]] where every processors are considered as workers and own parts of data. The [[SPMD|SPMD parallelism]] introduced in mid 1980 is particularly efficient in homogeneous computing environments for large data, for example, performing [[singular value decomposition]] on a large matrix, or performing [[Mixture model|clustering analysis]] on high-dimensional large data. On the other hand, there is no restriction to use [[Master/slave (technology)|manager/workers parallelism]] in [[SPMD|SPMD parallelism]] environment.

* The Rmpi<ref name=rmpi/> uses [[Master/slave (technology)|manager/workers parallelism]] where one main processor (manager) servers as the control of all other processors (workers). The [[Master/slave (technology)|manager/workers parallelism]] introduced around early 2000 is particularly efficient for large tasks in small [[Computer cluster|clusters]], for example, [[Bootstrapping (statistics)|bootstrap method]] and [[Monte Carlo method|Monte Carlo simulation]] in applied statistics since [[Independent and identically distributed random variables|i.i.d.]] assumption is commonly used in most [[Statistics|statistical analysis]]. In particular, task pull parallelism has better performance for Rmpi in heterogeneous computing environments.

The idea of [[SPMD|SPMD parallelism]] is to let every processors do the same works but on different parts of large data. For example, modern [[Graphics processing unit|GPU]] is a large collection of slower co-processors which can simply apply the same computation on different parts of relatively smaller data, but the SPMD parallelism ends up an efficient way to obtain final solutions, i.e. time to solution is shorter.<ref>{{cite web | url = http://graphics.stanford.edu/~mhouston/ | title = Folding@Home - GPGPU | author = Mike Houston | accessdate = 2007-10-04 }}</ref> It is clearly that pbdR is not only suitable for small [[Computer cluster|clusters]], but also is stabler for analyzing [[Big data]] and is more scalable for [[Supercomputer|supercomputers]].<ref>{{cite journal|author=Schmidt, D., Ostrouchov, G., Chen, W.-C., and Patel, P.|title=Tight Coupling of R and Distributed Linear Algebra for High-Level Programming with Big Data|year=2012|pages=811-815|journal=High Performance Computing, Networking, Storage and Analysis (SCC), 2012 SC Companion:|url=http://dl.acm.org/citation.cfm?id=2477156}}</ref> In short, pbdR

* does '''not''' like Rmpi, snow, snowfall, do-like, '''nor''' parallel packages in R,

* does '''not''' focus on interactive computing '''nor''' master/workers,

* but is able to use '''both''' SPMD and task parallelisms.

Two main implementations in [[R (programming language)|R]] using [[Message Passing Interface|MPI]] are Rmpi<ref name=rmpi>{{cite journal|author=Yu, H.|title=Rmpi: Parallel Statistical Computing in R|year=2002|url=httphttps://cran.r-project.org/package=Rmpi|journal=R News}}</ref> and pbdMPI of pbdR.

* The pbdR built on pbdMPI uses [[SPMD|SPMD parallelism]] where every processorsprocessor areis considered as workersworker and ownowns parts of data. The [[SPMD|SPMD parallelism]] introduced in mid 1980 is particularly efficient in homogeneous computing environments for large data, for example, performing [[singular value decomposition]] on a large matrix, or performing [[Mixture model|clustering analysis]] on high-dimensional large data. On the other hand, there is no restriction to use [[Master/slave (technology)|manager/workers parallelism]] in [[SPMD|SPMD parallelism]] environment.

* The Rmpi<ref name=rmpi/> uses [[Master/slave (technology)|manager/workers parallelism]] where one main processor (manager) serversserves as the control of all other processors (workers). The [[Master/slave (technology)|manager/workers parallelism]] introduced around early 2000 is particularly efficient for large tasks in small [[Computer cluster|clusters]], for example, [[Bootstrapping (statistics)|bootstrap method]] and [[Monte Carlo method|Monte Carlo simulation]] in applied statistics since [[Independent and identically distributed random variables|i.i.d.]] assumption is commonly used in most [[Statistics|statistical analysis]]. In particular, task pull parallelism has better performance for Rmpi in heterogeneous computing environments.

The idea of [[SPMD|SPMD parallelism]] is to let every processorsprocessor do the same worksamount of work, but on different parts of a large data set. For example, a modern [[Graphics processing unit|GPU]] is a large collection of slower co-processors whichthat can simply apply the same computation on different parts of relatively smaller data, but the SPMD parallelism ends up with an efficient way to obtain final solutions, (i.e. time to solution is shorter).<ref>{{cite web | url = http://graphics.stanford.edu/~mhouston/ | title = Folding@Home - GPGPU | author = Mike Houston | accessdateaccess-date = 2007-10-04 }}</ref> It is clearly that pbdR is not only suitable for small [[Computer cluster|clusters]], but also is stabler for analyzing [[Big data]] and is more scalable for [[Supercomputer|supercomputers]].<ref>{{cite journal|author=Schmidt, D., Ostrouchov, G., Chen, W.-C., and Patel, P.|title=Tight Coupling of R and Distributed Linear Algebra for High-Level Programming with Big Data|year=2012|pages=811-815|journal=High Performance Computing, Networking, Storage and Analysis (SCC), 2012 SC Companion:|url=http://dl.acm.org/citation.cfm?id=2477156}}</ref> In short, pbdR

== Package design ==

Line 31 ⟶ 29:

{| class="wikitable"

|-

! General !! I/O !! Computation !! Application !! Profiling !! Client/Server

|-

| pbdDEMO || pbdNCDF4 || pbdDMAT || pmclust || pbdPROF || pbdZMQ

|-

| pbdMPI || pbdADIOS || pbdBASE || pbdML || pbdBASEpbdPAPI || remoter

|-

| || || pbdSLAP || || hpcvis || pbdCS

|-

| || || kazaam || || || pbdRPC

|}

[[File:Pbd overview.png|thumb|The images describes how various pbdr packages are correlated.]]

Among these packages, pbdMPI provides wrapper functions to [[Message Passing Interface|MPI]] library, and it also produces a [[Library (computing)|shared library]] and a configuration file for MPI environments. All other packages rely on this configuration for installation and library loading that avoidavoids difficulty of library linking and compiling. All other packages can directly utilizeuse MPI functions easily.

* pbdMPI --- an efficient interface to MPI either [[Open MPI|OpenMPI]] or [[MPICH2]] with a focus on Single Program/Multiple Data ([[SPMD]]) parallel programming style

* pbdSLAP --- bundles scalable dense linear algebra libraries in double precision for R, based on [[ScaLAPACK]] version 2.0.2 which includes several scalable linear algebra packages (namely [[BLACS]], [[PBLAS]], and [[ScaLAPACK]]).

* pbdNCDF4 --- Interfaceinterface to Parallel Unidata [[NetCDF]]4 format data files

* pbdBASE --- low-level [[ScaLAPACK]] codes and wrappers

* pbdDMAT --- distributed matrix classes and computational methods, with a focus on linear algebra and statistics

* pbdDEMO --- set of package demonstrations and examples, and this unifying vignette

* pmclust --- parallel [[Mixture model|model-based clustering]] using pbdR

* pbdPROF --- profiling package for MPI codes and visualization of parsed stats

* pbdZMQ --- interface to [[ZeroMQ|ØMQ]]

* remoter --- R client with remote R servers

* pbdCS --- pbdR client with remote pbdR servers

* pbdRPC --- remote procedure call

* kazaam --- very tall and skinny distributed matrices

* pbdML --- machine learning toolbox

AmountAmong those packages, the pbdDEMO package is a collection of 20+ package demos which offer example uses of the various pbdR packages, and contains a vignette whichthat offers detailed explanations for the demos and provides some mathematical or statistical insight.

== Examples ==

=== Example 1 ===

Hello World! Save the following code in a file called ``"demo.r``"

<sourcesyntaxhighlight lang="rsplusr">

### Initial MPI

library(pbdMPI, quiet = TRUE)

Line 66 ⟶ 72:

### Finish

finalize()

</syntaxhighlight>

</source>

and use the command

<sourcesyntaxhighlight lang="bash">

mpiexec -np 2 Rscript demo.r

</syntaxhighlight>

</source>

to execute the code where [[R (programming language)|Rscript]]script is one of command line executable program.

=== Example 2 ===

The following example modified from pbdMPI illustrates the basic [[programming language syntax|syntax of the language]] of pbdR.

Since pbdR is designed in [[SPMD]], all the R scripts are stored in files and executed from the command line via mpiexec, mpirun, etc. Save the following code in a file called ``"demo.r``"

<sourcesyntaxhighlight lang="rsplusr">

### Initial MPI

library(pbdMPI, quiet = TRUE)

Line 95 ⟶ 101:

### Finish

finalize()

</syntaxhighlight>

</source>

and use the command

<sourcesyntaxhighlight lang="bash">

mpiexec -np 4 Rscript demo.r

</syntaxhighlight>

</source>

to execute the code where [[R (programming language)|Rscript]]script is one of command line executable program.

=== Example 3 ===

The following example modified from pbdDEMO illustrates the basic ddmatrix computation of pbdR which performs [[singular value decomposition]] on a given matrix.

Save the following code in a file called ``"demo.r``"

<sourcesyntaxhighlight lang="rsplusr">

# Initialize process grid

library(pbdDMAT, quiet=T)

Line 124 ⟶ 130:

# Finish

finalize()

</syntaxhighlight>

</source>

and use the command

<sourcesyntaxhighlight lang="bash">

mpiexec -np 2 Rscript demo.r

</syntaxhighlight>

</source>

to execute the code where [[R (programming language)|Rscript]]script is one of command line executable program.

== Further reading ==

* {{cite techreporttech report|author=Raim, A.M.|year=2013|title=Introduction to distributed computing with pbdR at the UMBC High Performance Computing Facility|institution=UMBC High Performance Computing Facility, University of Maryland, Baltimore County|number=HPCF-2013-2|url=http://wwwuserpages.umbc.edu/hpcf~gobbert/publicationspapers/indexpbdRtara2013.html#pdf|accessdate=2013-06-26|archiveurl=https://web.archive.org/web/20140204051402/http://userpages.umbc.edu/~gobbert/papers/pbdRtara2013.pdf|archivedate=2014-02-04|url-status=dead}}

* {{cite tech report|author=Bachmann, M.G., Dyas, A.D., Kilmer, S.C. and Sass, J.|year=2013|title=Block Cyclic Distribution of Data in pbdR and its Effects on Computational Efficiency|institution=UMBC High Performance Computing Facility, University of Maryland, Baltimore County|number=HPCF-2013-11|url=http://userpages.umbc.edu/~gobbert/papers/REU2013Team1.pdf|accessdate=2014-02-01|archiveurl=https://web.archive.org/web/20140204051351/http://userpages.umbc.edu/~gobbert/papers/REU2013Team1.pdf|archivedate=2014-02-04|url-status=dead}}

* {{cite web|title=High-Performance and Parallel Computing with R|author=Dirk Eddelbuettel|url=http://cran.r-project.org/web/views/HighPerformanceComputing.html}}

* {{cite tech report|author=Bailey, W.J., Chambless, C.A., Cho, B.M. and Smith, J.D.|year=2013|title=Identifying Nonlinear Correlations in High Dimensional Data with Application to Protein Molecular Dynamics Simulations|institution=UMBC High Performance Computing Facility, University of Maryland, Baltimore County|number=HPCF-2013-12|url=http://userpages.umbc.edu/~gobbert/papers/REU2013Team2.pdf|accessdate=2014-02-01|archiveurl=https://web.archive.org/web/20140204055902/http://userpages.umbc.edu/~gobbert/papers/REU2013Team2.pdf|archivedate=2014-02-04|url-status=dead}}

* {{cite news|title=R at 12,000 Cores|url=http://www.r-bloggers.com/r-at-12000-cores/}}<br />This article was read 22,584 times in 2012 since it posted on October 16, 2012 and ranked number 3<ref>{{cite news|url=http://www.r-bloggers.com/100-most-read-r-posts-for-2012-stats-from-r-bloggers-big-data-visualization-data-manipulation-and-other-languages/|title=100 most read R posts in 2012 (stats from R-bloggers) – big data, visualization, data manipulation, and other languages}}</ref>

* {{cite web|url=http://rwiki.sciviews.org/doku.php?id=developers:projects:gsoc2013:mpiprofiler|title=ProfilingHigh-Performance Tools forand Parallel Computing with R|author=GoogleDirk SummerEddelbuettel|date=13 ofNovember Code2022 |url=https://cran.r-project.org/web/views/HighPerformanceComputing.html|author-link=Dirk R 2013Eddelbuettel}}

* {{cite news|title=R at 12,000 Cores|url=http://www.r-bloggers.com/r-at-12000-cores/}}<br />This article was read 22,584 times in 2012 since it posted on October 16, 2012, and ranked number 3<ref>{{cite news|url=http://www.r-bloggers.com/100-most-read-r-posts-for-2012-stats-from-r-bloggers-big-data-visualization-data-manipulation-and-other-languages/|title=100 most read R posts in 2012 (stats from R-bloggers) – big data, visualization, data manipulation, and other languages}}</ref>

* {{cite web|url=http://rwiki.sciviews.org/doku.php?id=developers:projects:gsoc2013:mpiprofiler|archive-url=https://archive.today/20130629095333/http://rwiki.sciviews.org/doku.php?id=developers:projects:gsoc2013:mpiprofiler|url-status=dead|archive-date=2013-06-29|title=Profiling Tools for Parallel Computing with R|author=Google Summer of Code - R 2013}}

* {{cite web|url=http://rpubs.com/wush978/pbdMPI-linux-pilot|title=在雲端運算環境使用R和MPI|author=Wush Wu (2014)}}

* {{cite web|url=https://www.youtube.com/watch?v=m1vtPESsFqM|title=快速在AWS建立R和pbdMPI的使用環境|author=Wush Wu (2013)|website=[[YouTube]] }}

== References ==

{{Reflist|30em}}

== External links ==

* {{Official website|www.r-pbd.org}}

{{DEFAULTSORT:PbdR}}

[[Category:Parallel computing]]

[[Category:Programming languages]]

[[Category:Cross-platform free software]]

[[Category:FunctionalData languagesmining and machine learning software]]

[[Category:Data-centric programming languages]]

[[Category:Statistical software]]

[[Category:Free statistical software]]

[[Category:LinuxFunctional numerical analysis softwarelanguages]]

[[Category:DataNumerical mininganalysis andsoftware machinefor learning softwareLinux]]

[[Category:StatisticalNumerical analysis software for macOS]]

[[Category:Numerical analysis software for Windows]]

[[Category:Parallel computing]]