Exploiting Scientific Software to Solve Problems in Data Analytics

Karen Devine
Sandia National Laboratories

To model and simulate complex physical phenomenon in the nuclear weapon complex, the DOE has made significant investments in scientific software for high-performance computing. This talk explores ways the investments in the Trilinos solver framework have been exploited for applications in data science. Examples will include spectral clustering and sparse tensor decomposition.

Presentation (PDF File)

Back to Workshop III: HPC for Computationally and Data-Intensive Problems