Sparse hypermatrix cholesky: Customization for high performance

Ecient execution of numerical algorithms requires adapting the code to the underlying execution plat-form. In this paper we show the process of ne tun-ing our sparse Hypermatrix Cholesky factorization in order to exploit eciently two important machine re-sources: processor and memory. Using the techniques we presented in previous papers we tune our code on a dierent platform. Then, we extend our work in two directions: rst, we experiment with a variation of the ordering algorithm, and second, we reduce the data submatrix storage to be able to use larger sub-matrix sizes