Dear all:
I try a package called "rparallel", and it can make your multiple cores
desktop PC have parallel computation.
The following are my code:
########################
library(rparallel)
msd <- function(x)c(mean(x),sd(x))
m <- matrix(rnorm(100000*100),ncol=100)
st <- Sys.time()
M <- NULL
if( "rparallel" %in% names( getLoadedDLLs()) )
{
runParallel( resultVar="M", resultOp="rbind",nWorkers=2)
}else {
for(i in 1:100000){
M <- rbind(M, msd(m[i,]))
}
}
Sys.time() - st
########################
I know I can do the same thing by "apply". but just want to try the package.
and I got the following error message,
########################
...running parallel version
LCOM::read ERROR - No length read from pipe [\\.\pipe\worker0]
LCOM::receiveRequest ERROR - Reading Request Type from pipe [\\.\pipe\worker0]
LWM::manageWorker ERROR - Calling receiveRequest(). For worker: 0
########################
Does anyone have idea what happen?
My OS is windows XP sp3, and my cpu is Intel Pentium D 820 2.80GHz.
Thanks for every reply
¡×¡×¡×¡×¡×¡×¡×¡×¡×¡×¡×¡×¡×¡×¡×¡×
Oscar Chang ±i©v·¢
Research Assistant
Division of Biostatistics and Bioinformatics
National Health Research Institutes (NHRI)
Tel : 886-37-246-166 ext. 36116
email : [EMAIL PROTECTED]
[[alternative HTML version deleted]]
______________________________________________
R-help@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
