Cache-Oblivious algorithms are algorithms deisgned to take advantage of a CPU cache without having the size of the cache as an explicit parameter.
- Enclosed in this folder are two demonstrations of the speedups from utilizing cache-oblivious algorithms to improve performance.
- Complete with results, demos, and reports.
- Included is resources to follow up on any questions that may arrise.
The real experiment takes place in the Matrix Transpose directory. The Matrix multiply is more of a prequel experiment to demonstrate a simple version of isolating elements until they are small enough to be used by a cache. Matrix Transpose takes this to the next level and has a much more solid report explaining the results.