conditions. These conditions, which represent additional time-dependent partial differential equations, are ex-One of the great challenges in computational physics is the prediction of flow associated noise, where the quantities of interest, tremely important for successful aeroacoustic simulations.
namely the sound waves can be at high frequencies and are usually Thus computational techniques are challenged by the comorders of magnitude smaller in magnitude than the mean quantities. plexity of the physics, and the computational power is
In order to numerically resolve such small scales governed by the challenged by the required fine grids and large number of fluid dynamics equations, high resolution schemes are required.
computations.
Thus solutions of flow noise problems are computationally intensive. An efficient, hybrid, data parallel computational aeroacoustics Parallel computers offer high speeds and massive algorithm has been developed for the prediction of noise radiation amounts of memory, which make them attractive for and scattering from three-dimensional geometries. The algorithm computational aeroacoustic problems, where one has to solves the Euler/Navier-Stokes equations in the interior and nonresolve different sets of time dependent partial differential flecting boundary conditions on the outer boundaries. A moving equations; radiation boundary conditions on the inflow surface Kirchhoff method is coupled to the flow solver for far-field predictions. The algorithm uses standard time and spatial discretiza-boundaries, outflow boundary conditions on the outflow tion techniques but utilizes several new optimization strategies that boundaries [7,8], and the Euler/Navier-Stokes equations are highly suitable for single zone solutions on data parallel procesin the interior. Efficient handling of the outer boundaries is sors. One strategy, for example, enables simultaneous residual evala very difficult task when single zone solutions are peruations of the interior and far-field nonreflecting boundary condiformed in the data parallel mode on computers such as the tions equations, reducing the computational effort spent on them by approximately 60% CPU time savings. The algorithms for the flow Connection Machine-5 (CM-5). In the data parallel mode solver and the Kirchhoff method and their coupling are described in one solves for the interior points and the outer boundary this paper, and results for some example radiation and scattering points in a sequential manner. This, however, means at least problems are presented.