add toy example to demonstrate parallelisation

use %do% for single core
use %dopar% for multi core

parallel_example.r

+library(doParallel)
+registerDoParallel()
+
+x <- iris[which(iris[,5] != "setosa"), c(1,5)]
+trials <- 10000
+ptime <- system.time({
+    r <- foreach(icount(trials), .combine=cbind) %do% {
+          ind <- sample(100, 100, replace=TRUE)
+        result1 <- glm(x[ind,2]~x[ind,1], family=binomial(logit))
+            coefficients(result1)
+          }
+})[3]
+ptime