Parallelized Hybrid Merge Sorting Implementation Based On Ram Performance Using Mpich++

Main Article Content

Leta Tesfaye Jule, et. al.


Parallel handling alludes to the idea of accelerating the execution of a program by separating program into different sections that can execute all the while, each on its own processor. Consolidation sort is a proficient gap and-vanquish arranging calculation, consequently better comprehension of union sort parallelization can add to better comprehension of gap and-overcome parallelization by and large. This paper manages the execution of equal consolidation arranging method under MPI utilizing MPICH++for correspondence between the centers and for the calculation. Since it is particularly appropriate to execute in LINUX frameworks. The expert and the slave speak with one another utilizing MPICH++. In the calculation which we have carried out is for converge on a few hubs, it very well might be for just at least one slaves. One of equal cycles is assigned as an expert and remaining goes about as slaves. Unsorted rundown of components is gotten with randomized strategy. The expert partitions the unsorted rundown of components into the information parts equivalent to the quantity of slaves. We plan to assess and contrast these insights and the time taken to tackle a similar issue in sequential execution to show correspondence overhead engaged with equal calculation. We plan to analyze and assess the measurements acquired for various sizes of RAM under equal execution in a solitary hub including just two centers, where one goes about as expert and other as slave. Further shows reliance of sequential execution on RAM for similar issue by implementing its sequential form under different sizes of RAM. The proposed system results are analyzed in two phases whereas,  Serial Merge Sort with 256 MB RAM achieves 126.1472 peak range and 1000 MB RAM achieves2.0674with the time difference -124.0798sec. In the second phase parallel merge sorting with 256 MB RAM achieves 126.1472peak range and with 1000 MB RAM achieves 2.0674 with time difference -124.0798


Download data is not yet available.


Metrics Loading ...

Article Details

How to Cite
et. al., L. T. J. . (2021). Parallelized Hybrid Merge Sorting Implementation Based On Ram Performance Using Mpich++ . Turkish Journal of Computer and Mathematics Education (TURCOMAT), 12(12), 649–656.