Hi all,<br><br>I read about MPI_File_read_all and write correpondant, and the text says this operation perform reads using all processes defined in a certain group.<br><br>Well i have saw this operation is slower than file_read non collective version. Is it right?
<br><br>If i define 3 processes during mpirun, if i put this operation MPI_File_read_all in algorithm, this one will be executed 3 times, 3 times all processes will read all file?<br><br>Or it will read once a time? I recorded the time, and it shows three diferent times, then i think it have executed 3 times.
<br>The part of code is below<br><br> MPI_File_open(MPI_COMM_WORLD, "Wtupi.dat", MPI_MODE_RDONLY, MPI_INFO_NULL, &fh2);<br> MPI_File_set_view(fh2, 0, MPI_INT, MPI_INT, "native",MPI_INFO_NULL);
<br> start=MPI_Wtime();<br> MPI_File_read_all(fh2, buf2, 900000, MPI_INT, MPI_STATUS_IGNORE);<br> tempo=MPI_Wtime()-start;<br> printf("\nTime spent for read operation %f",tempo);<br> } <br><br>
Thanks<br>Luiz<br><br><br>