Nischay Mamidi - Academia.edu (original) (raw)
Uploads
Papers by Nischay Mamidi
ArXiv, 2021
This report presents a comprehensive analysis of the performance of GPU accelerated meshfree CFD ... more This report presents a comprehensive analysis of the performance of GPU accelerated meshfree CFD solvers for two-dimensional compressible flows in Fortran, C++, Python, and Julia. The programming model CUDA is used to develop the GPU codes. The meshfree solver is based on the least squares kinetic upwind method with entropy variables (q-LSKUM). To assess the computational efficiency of the GPU solvers and to compare their relative performance, benchmark calculations are performed on seven levels of point distribution. To analyse the difference in their run-times, the computationally intensive kernel is profiled. Various performance metrics are investigated from the profiled data to determine the cause of observed variation in run-times. To address some of the performance related issues, various optimisation strategies are employed. The optimised GPU codes are compared with the naive codes, and conclusions are drawn from their performance.
This paper presents the development of a Regent based implicitly parallel meshfree solver for inv... more This paper presents the development of a Regent based implicitly parallel meshfree solver for inviscid compressible fluid flows. The meshfree solver is based on the Least Squares Kinetic Upwind Method (LSKUM). The performance of the Regent parallel solver is assessed by comparing with the explicitly parallel versions of the same solver written in Fortran 90 and Julia. The Fortran code uses MPI with PETSc libraries, while the Julia code uses an MPI + X alternative parallel library. Numerical results are shown to assess the performance of these solvers on single and multiple CPU nodes.
ArXiv, 2021
This report presents a comprehensive analysis of the performance of GPU accelerated meshfree CFD ... more This report presents a comprehensive analysis of the performance of GPU accelerated meshfree CFD solvers for two-dimensional compressible flows in Fortran, C++, Python, and Julia. The programming model CUDA is used to develop the GPU codes. The meshfree solver is based on the least squares kinetic upwind method with entropy variables (q-LSKUM). To assess the computational efficiency of the GPU solvers and to compare their relative performance, benchmark calculations are performed on seven levels of point distribution. To analyse the difference in their run-times, the computationally intensive kernel is profiled. Various performance metrics are investigated from the profiled data to determine the cause of observed variation in run-times. To address some of the performance related issues, various optimisation strategies are employed. The optimised GPU codes are compared with the naive codes, and conclusions are drawn from their performance.
This paper presents the development of a Regent based implicitly parallel meshfree solver for inv... more This paper presents the development of a Regent based implicitly parallel meshfree solver for inviscid compressible fluid flows. The meshfree solver is based on the Least Squares Kinetic Upwind Method (LSKUM). The performance of the Regent parallel solver is assessed by comparing with the explicitly parallel versions of the same solver written in Fortran 90 and Julia. The Fortran code uses MPI with PETSc libraries, while the Julia code uses an MPI + X alternative parallel library. Numerical results are shown to assess the performance of these solvers on single and multiple CPU nodes.