Fig. 7From: Accelerating large-scale simulation of seismic wave propagation by multi-GPUs and three-dimensional domain decompositionScalability of multi-GPU computing with 3D decomposition. Each triplet value shows the size of the FDM domain. The complete scalability (proportional to the number of GPUs, N) and a scalability proportional to N2/3 are denoted by gray lines, respectively. For GPU, we used CUDA C and MPI library. For CPU, we parallelized the single-CPU Fortran code by using OpenMP and executed the program on a single host node.Back to article page