Practical aspect and experienceEvaluation of a spectral element CFD code on parallel architectures
References (13)
- et al.
Message passing on the Meiko CS-2
Parallel Comput.
(1994) The communication challenge for MPP: Intel Paragon and Meiko CS-2
Parallel Comput.
(1994)- et al.
Virtual Channel Router Version 2.0 User Guide
- et al.
Hardware Packet Router for VCR Version 2.0
Spectral element solution of the Navier-Stokes equations on high performance distributed-memory parallel processors
M.I.T. Thesis
(June 1989)- N. Floros, J. Reeve and O. Tutty. Δαιδαλoζ: Incompressible Navier-Stokes solver for parallel and distributed...
Cited by (5)
GPGPU implementation of mixed spectral-finite difference computational code for the numerical integration of the three-dimensional time-dependent incompressible Navier-Stokes equations
2014, Computers and FluidsCitation Excerpt :Briscolini [10] reported on three parallel message-passing implementations of three-dimensional pseudospectral Navier–Stokes solvers for homogeneous turbulence on three different IBM machines, one SP-1 and two SP-2 computers. Floros and Reeve [11] presented the evaluation of a spectral-element Navier–Stokes solver on three generations of parallel-computing architectures, namely a Inmos T-800 transputer, an Intel iPSC/860 and a Meiko CS-2. A mixed spectral-element, pseudospectral and finite-difference scheme for the Navier–Stokes equations has been implemented on a Meiko parallel computer by Prestin and Shtilman [12].
A parallel 3D unsteady incompressible flow solver on VPP700
2001, Parallel ComputingCitation Excerpt :Following Fourier pseudo-spectral method Pelz [5] worked on parallel numerical solution for NS equations on a 1024 noded hypercube and obtained an efficiency of 83%. Based on parallel spectral element method Fisher et al. [6], Floros and Reeves [7], Prestin and Shtilman [8], Tal [9], Passoni et al. [10], etc., have analysed NS equations on Intel hypercube, network of transputers, Intel ipsc/860, Meiko parallel computer, ALPHA workstations and CONVEX SPP 1200/XA, respectively. Briscolini and Fatica [11] developed a second-order finite difference NS solver on IBM SP2 and achieved an efficiency of 80–90%.
On direct numerical simulation of turbulent flows
2011, Applied Mechanics Reviews