In the codesign domain, many hardware and software techniques must be developed to satisfy specific constraints in terms of computation time, area, performance, power consumption, etc. This paper introduces an automatic approach To perform HW/SW partitioning and scheduling such that the global application execution time is minimized and the majority area of FPGA ( Field programmable gate array) used in RSoC ( Reconfigurable System on Chip) is exploited. The used algorithm is inspired by the collective behavior of social insects such as bees. : Honey Bees Mating Optimization (HBMO). Comparing the proposed method with Genetic algorithm, the simulation results show that the proposed algorithm has better convergence performance.

A parallel computing environment of an interconnected set of computers called MPI Cluster is set up on a Linux Operating System to reduce runtime of Density Functional Theory DFT calculations by combining the computational power of multiple computers. In this paper, we evaluate the performance of Quantum ESPRESSO (QE) on the MPI Cluster system. To test the speed and scalability of our cluster system, distinct-point sample work loads are being distributed over multiple MPI processors. The result implies that scaling speedup over many processors is only possible if the number of kpoints to parallelize is bigger than the number of processors. We also discovered that the speedup limit for parallelizing band calculations is somewhat independent of the number of bands employed and that it reduces linearly as the number of MPI processors increases.