Empirical Investigation Of Cloud, Grid and Virtualization Using Compilar Optimization Level For Threads Processe

Abstract

This research focused on the  implementation of Open MP. It considers the parallelization  of an application code that simulates the thermal gradient of  material in two dimensions. A ‘C language’ program code  called jacobi2d.c that solves a rectangular 2-dimensional  heat conductivity problem using Jacobi iterative method  was used. The boundary conditions required to compute a  temperature distribution for a rectangular 2D problem are:  Top 300C, Bottom 500C, Left 400C and Right 900C with a  range of problem sizes enter as a run-time parameter to alter  the problem sizes and convergence criteria. Also, there  were computations and readings for iterations and runtime  for four values of M and N which were selected for 01, 02,  and 03 optimizations. In Table 1 Readings, four values  were selected for each of the iterations. The results show  the performance of the runtime as the processor increases  from 01-optimization, to 02-optimization and finally to 03- optimization. It can be deduced from the representation that  the run time of the values reduces as more resources are  allocated to execution through the increase in optimization  level. Also, in Table 2 Readings, the runtime decreases as it  moves from thread1, thread2, thread3, and thread4,  comparing the last values for thread1 which are M is 180, N  is 200, and their runtime which is 42.797187001. Also the  last values for thread2 which are M is 180, N is 200, their  runtime which is 21.772106003. When the two runtimes  were compared, it was discovered that there was a decrease  in the runtime because the more the thread increases, the  more system resources they share such as a processor which  may affect their runtime by increasing it.