我正在使用C和MPI开发nBody模拟,并将其部署在AWS EC2集群上。 我已经创建了一个bash脚本来在集群上运行我强大的扩展测试用例
#!/bin/bash
mpirun -np 1 --hostfile hostfile.txt main2 50000 20 >> test50kX20np1.txt
echo "Test 1 Terminato"
mpirun -np 2 --hostfile hostfile.txt main2 50000 20 >> test50kX20np2.txt
echo "Test 2 terminato"
mpirun -np 4 --hostfile hostfile.txt main2 50000 20 >> test50kX20np4.txt
echo "Test 4 terminato"
mpirun -np 6 --hostfile hostfile.txt main2 50000 20 >> test50kX20np6.txt
echo "Test 6 terminato"
mpirun -np 8 --hostfile hostfile.txt main2 50000 20 >> test50kX20np8.txt
echo "Test 8 terminato"
mpirun -np 10 --hostfile hostfile.txt main2 50000 20 >> test50kX20np10.txt
echo "Test 10 terminato"
mpirun -np 12 --hostfile hostfile.txt main2 50000 20 >> test50kX20np12.txt
echo "Test 12 terminato"
mpirun -np 14 --hostfile hostfile.txt main2 50000 20 >> test50kX20np14.txt
echo "Test 14 terminato"
mpirun -np 16 --hostfile hostfile.txt main2 50000 20 >> test50kX20np16.txt
echo "Test 16 terminato"
在脚本的末尾生成了txt:
test50kX20np2.txt,
test50kX20np4.txt,
他们有这一行:(例如在 test50kX20np4.txt 中)
4 total processes killed (some possibly by mpirun during cleanup)
可能是由于此错误导致的?