This study presented a substructure based parallel linear solution framework for the static analysis of linear structural engineering problems having multiple loading conditions. The framework was composed of two separate programs designed to work on PC Clusters having the Windows operating system. The first program was responsible for creating the optimum substructures for the parallel solution and first partitioned the structure in such a way that the number of substructures was equal to the number of processors. Then, the estimated condensation time imbalance of the initial substructures was adjusted by iteratively transferring nodes from the substructures with slower estimated condensation times to the substructures with faster estimated condensation times. Once the final substructures were created, the second program started the solution. Each processor assembled its substructures stiffness matrix and condensed it to the interface with other substructures. The interface problem was solved by a parallel variable band solver. After computing the interface unknowns, each processor calculated the internal displacements and element stresses or forces. Examples which illustrate the applicability and efficiency of this approach were also presented. In these examples, the number of processors was varied from one to twelve to demonstrate the performance of the overall solution framework.

The work deals with the encoding and decoding of Reed-Solomon codes. There is generally described algebraic decoding of Reed-Solomon codes, and then described four methods of decoding, namely Massey-Berlekamp algorithm, Euclidean algoritus, Peterson-Gorenstein-Zierler algorithm and the direct method. These methods are then compared, and some of them are implemented in Matlab.