Parallelization in R: %dopar% vs %do%. Why using a single core yields to better performance?

Parallelization in R: %dopar% vs %do%. Why using a single core yields to
better performance?

References

Reference NO 1

http://stackoverflow.com/questions/15194465/parallelization-in-r-dopar-vs-do-why-using-a-single-core-yields-to-better

Reference NO 2

http://stackoverflow.com/questions/18257090/parallel-computing-in-r-windows-changing-code-from-foreach-do-to-foreach-d

Reference NO 3

http://stackoverflow.com/questions/16650774/using-all-cores-for-r-massstepaic-process

Reference NO 4

http://stackoverflow.com/questions/16219228/is-it-possible-in-r-parallelmcparallel-to-limit-the-number-of-cores-used-at

Reference NO 5

http://www.slideshare.net/bytemining/taking-r-to-the-limit-high-performance-computing-in-r-part-1-parallelization-la-r-users-group-727

Reference NO 6

http://nimbios.org/tutorials/talks/Pragnesh-talk.pdf

Reference NO 7

#EANF#

Reference NO 8

#EANF#