Shared memory scalar parallel computer for parallel processing method tridiagonalization real symmetric matrix

When performing tridiagonalization real symmetric matrix in shared memory scalar parallel computer, and updates the matrix by dividing each thread. In this case, the matrix is ​​because it is symmetric matrix, calculated using the data of the lower triangular matrix portion. Lower triangular matrix portion is divided into vertical, but allocated to each CPU, this time, the number of elements to be allocated to each CPU allocated to be the same. Then, once loading data, performs vertical calculation and transverse simultaneous operations, reduce the number of memory load. Lower triangular matrix portion of the diagonal block matrix portion, the updated while recursively performs be divided into one small square matrix and two small lower triangular matrix.