Abstract: BLAS level3 subprograms have high computation complexity, which usually become applications' performance bottleneck. By organizing largescale floatingpoint units into a linear array architecture, the matrix multiplication coprocessor can perform highperformance and efficient matrix multiplication. Achieving efficient BLAS level3 computation on the matrix multiplication coprocessor is essential for the acceleration of largescale science and engineering applications. 
By taking matrix multiplication as the kernel and combining the characteristics of the underlying linear array architecture, this paper proposes the design of BLAS level3 computation on a matrix multiplication coprocessor, and construct a corresponding performance model. Experimental results show that SYMM, SYRK and TRMM subprograms on the matrix multiplication coprocessor achieves the computation efficiency of 99%, 98% and 80% respectively, at most 31% higher than those on the SW26010 and NVIDIA V100 GPU.

Key words: linear array, matrix multiplication, coprocessor, BLAS level-3