PARALELL STAT COMPUTING - Simple Parallel Statistical...

Info iconThis preview shows pages 1–9. Sign up to view the full content.

View Full Document Right Arrow Icon
Simple Parallel Statistical Computing in R Luke Tierney University of Iowa Joint work with A. J. Rossini and Na Li Biostatistics, University of Washington March 13, 2003
Background image of page 1

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Simple Parallel Statistical Computing in R March 13, 2003 What Is R? R is a language for statistical computing and graphics. Similar to S (John Chambers et al., Bell Labs). ACM Software System Award, 1999. De facto standard for computing in Statistical research. Documented in many books, e.g. Venables and Ripley. Can view R as a different implementation or dialect of S. There are some important differences, but much code written for S runs unaltered under R. 1
Background image of page 2
Simple Parallel Statistical Computing in R March 13, 2003 History and Development Model R is an Open Source project. Originally developed by Robert Gentleman and Ross Ihaka. Developed by the R-core group since mid 1997, Douglas Bates John Chambers Peter Dalgaard Robert Gentleman Kurt Hornik Stefano Iacus Ross Ihaka Friedrich Leisch Thomas Lumley Martin Maechler Guido Masarotto Paul Murrell Brian Ripley Duncan Temple Lang Luke Tierney 2
Background image of page 3

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Simple Parallel Statistical Computing in R March 13, 2003 Why Parallel Computing? Many computations seem instantaneous. Some would take hours, days, or months. Often multiple processors are available: multiple workstations dedicated cluster high-end SMP machine Can we make effective use of these resources? 3
Background image of page 4
Simple Parallel Statistical Computing in R March 13, 2003 Ideal Performance Improvement p processors should be p times faster than one processor. Some time scales: Single processor 30 Processors 1 minute 2 seconds 1 hour 2 minutes 1 day 1 hour 1 month 1 day 1 year 2 weeks 4
Background image of page 5

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Simple Parallel Statistical Computing in R March 13, 2003 Ideal Programming Requirement Minimal effort for simple problems. Be able to use existing high level (i.e. R) code. Ability to test code in sequential setting. 5
Background image of page 6
Simple Parallel Statistical Computing in R March 13, 2003 Parallel Computing on Networks of Workstations Use multiple cooperating processes. One process per available processor. Processes need to communicate with each other. Usually one process communicates with the user. 6
Background image of page 7

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Simple Parallel Statistical Computing in R March 13, 2003 Available Communications Mechanisms Sockets Message passing libraries (PVM, MPI) very powerful not easy to use designed for C, FORTRAN R interfaces socket connections rpvm, Rmpi 7
Background image of page 8
Image of page 9
This is the end of the preview. Sign up to access the rest of the document.

This note was uploaded on 05/12/2010 for the course APPLIED ST 2010 taught by Professor Various during the Spring '10 term at Universidad Nacional Agraria La Molina.

Page1 / 31

PARALELL STAT COMPUTING - Simple Parallel Statistical...

This preview shows document pages 1 - 9. Sign up to view the full document.

View Full Document Right Arrow Icon
Ask a homework question - tutors are online