Fluid-Structure Interaction (FSI) occurs in many engineering applications. One of the applications is structural acoustic analysis for submerged structures (e.g., UUV, Wind Turbine) subjected to external excitations. To better understand the leading sources of vibration and sound, it requires knowledge of various disciplines (e.g., structural dynamics, acoustics, and wave propagation) and their interactions. Numerical methods have been widely used to conduct structural acoustic analysis. However, as the numerical models are getting more complex and larger (degree of freedom), conventional approaches take too long to complete analyses and to support design optimization. This paper will discuss two advanced numerical techniques along with High Performance Computing (HPC) to effectively solving the FSI analysis. The first numerical technique is “Finite Element Tearing and Interconnecting (FETI)” which is an iterative and a massively parallel solver developed by researchers from Stanford university. It can effectively scale up across hundreds and thousands computing nodes. The other technique is “Adaptive Krylov subspace and Galerkin Projection (AKGP)” which applies user-defined tolerance to speed-up frequency-sweep of a FE model. Instead of calculating results at every frequency point, the AKGP only calculates a small subset of the original frequency range and use AKGP with the defined tolerance to predict results for the rest of frequency points. Couple examples will be presented to show the effectiveness of combing the two advanced techniques with HPC in conducting the structural acoustic predictions. Especially, a large FE model with more than 60M Degree of Freedom (DOF) is exercised to demonstrate the techniques’ scalability and total turn-around time. Furthermore, the influences of the user-defined tolerance (error estimate) on efficiency and accuracy of the FSI prediction will be presented. REFERENCES [1] C. Farhat and F. X. Roux, A method of finite element tearing and interconnecting and its parallel solution algorithm. Internat. J. Numer. Meths. Engrg. Vol. 32, pp. 1205-1227, 1991 [2] P. Avery, C. Farhat, G. Reese, Fast frequency sweep computations using a multi-point Padé-based reconstruction method and an efficient iterative solver. Internat. J. Numer. Meths. Engrg. Vol. 69, pp. 2848-2875, 2007.