Tiling (folding)
Beginner
Cutting a big job into pieces the size of the grid and running them one after another.
Novice
Splitting a matrix multiplication that is bigger than the array into blocks (tiles) that fit, running each, and combining the results. Edge tiles may only partly fill the array.
Expert
A GEMM mapped onto an array needs folds. Tile order sets how often each operand is re-read from the buffer; tile size relative to the matrix sets the fragmentation loss.