Or better yet, write a compute shader since hashing is an embarrassingly parallel operation.
That said, the OP's article is correct in that straightforward idiomatic implementations of this algorithm are very much compute bound. The corollary is that eng work put into optimizing compute usage often won't be waisted for programs processing disk data (or even network data with modern 10Gb fiber connections).
That said, the OP's article is correct in that straightforward idiomatic implementations of this algorithm are very much compute bound. The corollary is that eng work put into optimizing compute usage often won't be waisted for programs processing disk data (or even network data with modern 10Gb fiber connections).