PySpark code that calculates A x A^T x A. Tested with a 1,000,000 x 1000 matrix in a computing cluster. The included file data-small-sample.txt contains space-separated floating point values that make up a 1,000 x 200 matrix for testing.
jaakko-paavola/massive-matrix-multiplication-with-spark
PySpark code that calculates A x A^T x A distributed in a computing cluster.
Jupyter NotebookMIT