MPI Scalability of a Large Memory LES Code
نویسندگان
چکیده
Large eddy simulations (LES) of co-flowing round jets with 33.6 million grid points are carried out using 16 Sun 1 GHz UltraSPARC III Cu processors. An in-depth processor scalability analysis is carried out for an MPI based code for a finite-volume solution of time dependent Navier-Stokes equations. The solver is based on a second-order structured staggered grid discretization, second-order time advancement, and a multi-grid Poisson equation for pressure. Particular attention is paid to the effect of initial conditions on the spatial development of the co-flowing jet at a Reynolds number of 7,300. A co-flow velocity to initial jet centreline velocity ratio of 1:11 and a coflow to initial jet diameter ratio of 35:1 are used to match the flow cases of [12]. The simulation volume, where is the orifice diameter at the jet inlet, is divided into control volumes in the longitudinal, radial and azimuthal directions respectively. Time averaged results of the effect of initial conditions on mean flow, the decay of jet centreline velocity, growth of the jet and the distribution of Reynolds stresses in the near and far field of the shear layer are presented. These quantities show good agreement with the measurements of [12]. The evolution of the turbulence structures is quantified using various discrimination techniques, and when compared with the flow visualisation experiments of [15], both techniques are observed to extract similar topological features.
منابع مشابه
A Message-Passing Distributed Memory Parallel Algorithm for a Dual-Code Thin Layer, Parabolized Navier-Stokes Solver
In this study, the results of parallelization of a 3-D dual code (Thin Layer, Parabolized Navier-Stokes solver) for solving supersonic turbulent flow around body and wing-body combinations are presented. As a serial code, TLNS solver is very time consuming and takes a large part of memory due to the iterative and lengthy computations. Also for complicated geometries, an exceeding number of grid...
متن کاملScalable Quasineutral Solver for Gyrokinetic Simulation
Modeling turbulent transport is a major goal in order to predict con nement issues in a tokamak plasma. The gyrokinetic framework considers a computational domain in ve dimensions to look at kinetic issues in a plasma. Gyrokinetic simulations lead to huge computational needs. Up to now, the gyrokinetic code GYSELA performed large simulations using a few thousands of cores. The work proposed her...
متن کاملScalable Memory Use in MPI: A Case Study with MPICH2
One of the factors that can limit the scalability of MPI to exascale is the amount of memory consumed by the MPI implementation. In fact, some researchers believe that existing MPI implementations, if used unchanged, will themselves consume a large fraction of the available system memory at exascale. To investigate and address this issue, we undertook a study of the memory consumed by the MPICH...
متن کاملParallel Stratiied and Rotating Turbulence Sim- Ulation Based on Mpi
We describe a parallel implementation for large-eddy simulation(LES) of the strati ed and rotating turbulence based on MPI. The parallelization strategy is speci ed by eliminating the tridiagonal solver with explicit method and by a domain decomposition for solving the poisson equation. In this simulation we have run on CRAY-T3E under the message passing platform MPI with a various domain decom...
متن کاملLarge Eddy Simulation of Turbulent Flows Using the CFD Code GASFLOW - MPI
GASFLOW-MPI is a scalable CFD software solution used to predict fluid dynamics, conjugate heat and mass transfer, chemical kinetics, aerosol transportation and other related phenomena. The generalized 3-D transient, compressible Navier-Stokes equations for multispecies are solved. It has been widely used to analyze safety problems in nuclear engineering. Turbulence models based on the ReynoldsA...
متن کامل