Abstract
Amount of data is rising rapidly with the passage of time. Sorting is well known computer science problem that is widely used in various applications. So there is need of certain sorting techniques that can arrange this data as fast as possible. The analysis of sorting algorithms on performance basis has significance in understanding that which technique is most effectual in the field of data management and which approach can arrange massive data accurately in least duration of time. The performance of an algorithm is measured on the basis of time complexity, space complexity and computation complexity. Multicore computer architecture attracts the researchers towards parallel computing for attaining highest computational performance from computer systems. In this research paper, an experimental analysis is conducted to measure the performance of On(log n) class sorting algorithms in terms of execution time in parallel manner Only same On(log n) class 12 algorithms are analyzed that leads this work towards novel results. Experimentation is performed using C++ language and OpenMP library is implemented for standard parallelism. Data size increase in terms of 2 power N. Test cases are executed with three type of following integer data; random integers, sorted integers and reversed sorted integers. State of the art results are illustrated using comparative graphs that shows the performance of different algorithms under same scenario. This research work help to select appropriate sorting technique with regard to data set and environment.