Performance Analysis of Parallelizing Program Code for Multi-Core Architectures Using OPENMP
Multi-core architectures have become more popular due to better performance, reduced heat dissipation, power consumption and more efficient simultaneous processing of multi tasks. If one wishes to run a single application faster, then that application must be divided into sub programs, or threads, that cooperate to deliver the desired functionality. The Open MP programming paradigm implements loop level parallelism, which is one of the most basic available units of parallelism for parallel Open MP programs. Loop-level parallelism allows an Open MP implementation to easily split the work across multiple threads. This paper mainly focuses on Parallelizing programming code using Open MP and analyzes the program performance using V-tune.