This paper presents an implementation of different matrix-matrix multiplication routines in OpenCL. We utilize the high-performance GEMM (GEneral Matrix-Matrix Multiply) implementation from our pre...