Abstract: To reduce the parallel efficiency degradation of the system dynamic analysis on an entire large structure caused by the heterogeneous multi-core and distributed memory parallel computers, a hybrid hierarchical parallel algorithm for structure dynamic analysis is proposed on the basis of distributed domain solver. In this algorithm, the distributed storage of large amount of data is introduced to improve the memory access. To make up for the shortcoming that the communication cost increases with increase of the interface equation size when using the traditional distributed domain solver, a hybrid partitioning based on distributed domain solver and parallel solver is introduced to reduce the interface equation size and further reduce the solution time. Moreover, a three-layer parallelization of the computational procedure is introduced to enable the separation of the communication of inter-node, heterogeneous core groups and inside- heterogeneous core-groups, which can significantly improve the communication rate. Thus, it can improve the efficiency rates of parallel computing of large-scale dynamic analysis by fully exploiting the architecture characteristics of the heterogeneous multi-core and distributed memory parallel computers. By solving benchmark instances, the effectiveness and superiority of the proposed method are proved.