From: Block size estimation for data partitioning in HPC applications using machine learning techniques
Symbol | Meaning |
---|---|
d | Dataset composed by n rows and m columns |
e | A representation of the target execution environment |
a | The machine learning algorithm to be executed on d within e |
\(p_r^*\) | The optimal number of partitions along rows |
\(p_c^*\) | The optimal number of partitions along columns |
\(r^*\) | The optimal number of rows in a block |
\(c^*\) | The optimal number of columns in a block |
\((r^*, c^*)\) | The optimal block size expressed as \((n/p_r^*, m/p_c^*)\) |