EFFECTIVENESS OF OPENMP FOR ALGORITHM PARALLELIZATION

Today in the market, highly efficient, scalable and fast processors are available. This was all about the hardware perspective. But the software markets have not scaled up in the similar fashion. To scale up software efficiency, OpenMP tried to offer a shared memory parallel programming model. OpenMP (Open Multiprocessing) is an Application Program Interface (API) that can be used to explicitly direct multi-threaded, shared memory parallelism. OpenMP is not a new computer language; it works in conjunction with either standard FORTRAN or C/C+ +. This paper illustrates the basic concepts of parallel computing with a brief overview of OpenMP. The paper also describes an analysis of algorithms from different fields like LU decomposition, PI Value using Monte Carlo method. The observations and results obtained show that how the usage of OpenMP's Pragma are effective in the normal C/C++ programs and how the result varies according to the inputs and available number of threads and shows that it is useful only when we are working on large data set or large computations are involved in the given problem.