an important research problem (Rajaraman and Ullman, 2011; Wu
et al., 2014; Zhou et al., 2014).
Mining with big data or big data mining is very hard to manage
using the current methodologies and data mining software
tools due to their large size and complexity (Fan and Bifet, 2012).
In other words, using a single personal computer (PC) to execute
the data mining task over large scale datasets requires very
high computational costs. It is necessary to use more powerful
computing environments to efficiently process and analyze big
data.
According to Wu et al. (2014), the solutions for the problem of
mining large scale datasets can be based on the parallel and cloud
computing platforms. In principle, parallel computing focuses on
dividing the chosen (large) problem into smaller ones, each of
which (i.e. calculation) is carried out by one single processor individually,
so that a computation composed of a number of calculations
is performed concurrently in a distributed and parallel manner
(Gottlieb and Almasi, 1989). This leads to some research issues