Financial institutions stand at the edge of what is called a dynamic environment. Acting fast is a vital requirement of this environment. On the top of that, changes at the regulations [31, 32] make institutions' obligations more complex and compute intensive than ever. Financial researchers try constantly to create new models and improve the existing ones to reduce the complexity and make them simpler and more efficient. In addition, computer scientists have made a significant progress in the acceleration of complex algorithms using parallel and distributed systems. High Performance Computing techniques are used more often to meet the speed requirements that have been set by the modern economic situations. In this thesis we will present a state-of-the-art financial method, called Least-Squares Monte Carlo. This method is presented to reduce the complexity of a more complex method called Stochastic-on-Stochastic valuation. New regulations have introduced by the European Commission for the insurance companies and this algorithm can fulfill the requirements of these regulations faster and accurately. Furthermore, the utilization of the High Performance Computing techniques will reveal the power of the modern hardware architectures and programming languages as they can be able to accelerate the performance of a complex financial application by multiple times.