Parallel processing is a form of computing in which number of activities are carried out concurrently on multiple machines so that the time required to solve a given problem is reduced. Several interesting programming models have been developed for parallel computing applications such as OpenMP for shared memory programming or multicore machines and MPI (Message Passing Interface) for distributed memory network of workstations.
For decades, researchers have been using high-performance Personal Computers (PC) and open source LINUX operating system-based clusters for parallel computing experiments using MPI. Recently the phenomenal popularity of low cost and power efficient single board multicore computers like the Raspberry Pi has caused the emergence of the single board computer based Micro Clusters. Micro-clusters have greater scalability, power efficiency and shortdevelopment time. In this talk, our experience of developing a five node Raspberry Pi based
Micro cluster will be presented along with simple examples of MPI applications implemented on the cluster