FFT_SERIAL is a C program which demonstrates the computation of a Fast Fourier Transform, and is intended as a starting point for developing a parallel version using OpenMP.
On an Apple PowerPC G5 with two processors, the following results were observed:
| N | Time | MegaFLOPS |
|---|---|---|
| 2 | 4.943500e-07 | 20.2 |
| 4 | 1.217550e-06 | 32.8 |
| 8 | 2.283850e-06 | 52.5 |
| 16 | 5.334400e-06 | 59.9 |
| 32 | 1.051800e-05 | 76.0 |
| 64 | 2.125550e-05 | 90.3 |
| 128 | 3.229850e-05 | 138.7 |
| 256 | 7.576800e-05 | 135.1 |
| 512 | 1.544650e-04 | 149.1 |
| 1024 | 3.537100e-04 | 144.7 |
| 2048 | 7.222600e-04 | 155.9 |
| 4096 | 1.636675e-03 | 150.1 |
| 8192 | 3.392550e-03 | 156.9 |
| 16384 | 7.518900e-03 | 152.5 |
| 32768 | 1.601240e-02 | 153.4 |
| 65536 | 3.534380e-02 | 148.3 |
| 131072 | 7.358900e-02 | 151.3 |
| 262144 | 1.712250e-01 | 137.78 |
| 524288 | 3.547840e-01 | 140.3 |
| 1048576 | 7.744890e-01 | 135.3 |
BT_SERIAL is a FORTRAN90 program which is a serial version of the NAS Parallel Benchmark BT (block tridiagonal solver for a 3D PDE).
CG_SERIAL is a FORTRAN77 program which is a serial version of the NAS Parallel Benchmark CG (conjugate gradient solver).
EP_SERIAL is a FORTRAN77 program which is a serial version of the NAS Parallel Benchmark EP (embarassingly parallel).
FFT_OPEN_MP is a C program which computes a Fast Fourier Transform using OpenMP.
FFT_SERIAL is available in a C version and a C++ version and a FORTRAN77 version and a FORTRAN90 version and a MATLAB version.
HEATED_PLATE is a C program which solves the steady state heat equation in a 2D rectangular region, and is intended as a starting point for implementing an OpenMP parallel version.
IS_SERIAL is a C program which is a serial version of the NAS Parallel Benchmark IS (integer sort).
MD is a C program which carries out a molecular dynamics simulation, and is intended as a starting point for developing a parallel version using OpenMP.
OPEN_MP is a directory of C examples which illustrate the use of the OpenMP application program interface for carrying out parallel computations in a shared memory environment.
You can go up one level to the C source codes.