基于消息传递接口(Message Passing Interface,MPI),用两种不同的并行程序设计方法对Autodock程序进行修改.将修改后的程序应用于HIV-1蛋白酶(Protease)和小分子抑制剂XK263的对接体系,测试了并行程序的加速比和并行效率.结果表明,两种改进的并行Autodock程序都可以很好地完成计算,尤其是方案Ⅱ并行程序的加速比和并行效率更高.
Based on message passing interface (MPI), Autodock code is modified by two parallel methods and applied to dock small molecules XK263 to target HIV-1 protease .The acceleration rate and parallel efficiency are tested with different number of nodes. The improved Autodock codes show a satisfied performance during docking process. Particularly, Scheme Ⅱ has advantages in acceleration rate and parallel efficiency.