Jump to: navigation, search

GPU621/Threadless Horsemen

362 bytes added, 13:47, 26 November 2018
OpenMP vs Julia Results
* The reason loop interchange works for OpenMP is the way we store our array in memory originally favored "Row-Major" which allows the processor to move across cached data in a row fashion faster than column based... * As you might have seen, Julia’s Loop loop interchange is worseit's for opposite reason OpenMP improves from loop interchange.*[ Julia favours "Column-Major" layouts in cache memory.] [[File:255px-Row_and_column_major_order.svg.png]]
* Julia has several levels of runtime optimization (0-3)
* julia -O2 scriptName.jl or julia --optimize=2
* Set the optimization level (default level is 2 if unspecified or 3 if used without a level -O)
* recap loop interchange benefits for openmp (locality of reference)* discuss julia storing arrays as column major, loop interchange was worse for julia* discuss different levels of optimizationvectorization...
== Conclusion ==

Navigation menu