If I have a char array that represents, for example, integer value(s), and I used it to send these values via MPI with the appropriate MPI datatype for send and receive operations as follows:
int main(int argc, char* argv[]){
int my_rank; /* rank of process */
MPI_Comm_rank(MPI_COMM_WORLD, &my_rank);
unsigned char buff[100];
if (my_rank == 0){
int n = 99;
int i;
for(i = 0; i < sizeof(n); i++){
buff[i] = (n >> (8 * i)) & 0xFF;
}
MPI_Send(&buff, 1, MPI_INT, 1, 0, MPI_COMM_WORLD);
}
else{
MPI_Recv(&buff, 1, MPI_INT, 0, 0, MPI_COMM_WORLD, NULL);
int n = *(int *)buff;
printf("%d\n", n);
}
/* shut down MPI */
MPI_Finalize();
return 0;
}
Is this portable across machines of different architecture/endiness?
I suspect the only part that is not portable is the conversion from integer value to char array:
int i;
for(i = 0; i < sizeof(n); i++){
buff[i] = (n >> (8 * i)) & 0xFF;
}
But, anyways if not, is there a way to make the above program fully portable with the existence of the char array to store value(s)?
The way you serialize integers into a buffer is not portable. But if you are sending the integer, why not send integers directly, just take endianness into account. Suppose you want to sent 32bits integers:
int n = 99;
int sent = htonl(n);
MPI_Send(&send, 1, MPI_INT, 1, 0, MPI_COMM_WORLD);
For receiving:
int n;
int recv;
MPI_Recv(&recv, 1, MPI_INT, 0, 0, MPI_COMM_WORLD, NULL);
n = ntonl(recv);
Related
This is the program I am using to sum all values in a 1D array, and it works correctly. But how do I modify it to work on 2D array? Imagine variable a is something like a = { {1,2}, {3,4}, {5,6} };.
I tried few solutions but they are not working, so can someone explain few important changes to make to make it compatible with 2D array also.
#include <mpi.h>
#include <stdio.h>
#include <stdlib.h>
// size of array
#define n 10
int a[] = { 1, 2, 3, 4, 5, 6, 7, 8, 9, 10 };
// Temporary array for slave process
int a2[1000];
int main(int argc, char* argv[])
{
int pid, np,
elements_per_process,
n_elements_recieved;
// np -> no. of processes
// pid -> process id
MPI_Status status;
// Creation of parallel processes
MPI_Init(&argc, &argv);
// find out process ID,
// and how many processes were started
MPI_Comm_rank(MPI_COMM_WORLD, &pid);
MPI_Comm_size(MPI_COMM_WORLD, &np);
// master process
if (pid == 0) {
int index, i;
elements_per_process = n / np;
// check if more than 1 processes are run
if (np > 1) {
// distributes the portion of array
// to child processes to calculate
// their partial sums
for (i = 1; i < np - 1; i++) {
index = i * elements_per_process;
MPI_Send(&elements_per_process,
1, MPI_INT, i, 0,
MPI_COMM_WORLD);
MPI_Send(&a[index],
elements_per_process,
MPI_INT, i, 0,
MPI_COMM_WORLD);
}
// last process adds remaining elements
index = i * elements_per_process;
int elements_left = n - index;
MPI_Send(&elements_left,
1, MPI_INT,
i, 0,
MPI_COMM_WORLD);
MPI_Send(&a[index],
elements_left,
MPI_INT, i, 0,
MPI_COMM_WORLD);
}
// master process add its own sub array
int sum = 0;
for (i = 0; i < elements_per_process; i++)
sum += a[i];
// collects partial sums from other processes
int tmp;
for (i = 1; i < np; i++) {
MPI_Recv(&tmp, 1, MPI_INT,
MPI_ANY_SOURCE, 0,
MPI_COMM_WORLD,
&status);
int sender = status.MPI_SOURCE;
sum += tmp;
}
// prints the final sum of array
printf("Sum of array is : %d\n", sum);
}
// slave processes
else {
MPI_Recv(&n_elements_recieved,
1, MPI_INT, 0, 0,
MPI_COMM_WORLD,
&status);
// stores the received array segment
// in local array a2
MPI_Recv(&a2, n_elements_recieved,
MPI_INT, 0, 0,
MPI_COMM_WORLD,
&status);
// calculates its partial sum
int partial_sum = 0;
for (int i = 0; i < n_elements_recieved; i++)
partial_sum += a2[i];
// sends the partial sum to the root process
MPI_Send(&partial_sum, 1, MPI_INT,
0, 0, MPI_COMM_WORLD);
}
// cleans up all MPI state before exit of process
MPI_Finalize();
return 0;
}
You can simplify a lot by using MPI_Reduce instead of MPI_Send/MPI_Recv:
Reduces values on all processes to a single value
A nice tutorial about that routine can be found here.
So each process contains an array (e.g., process 0 { 1, 2, 3, 4, 5} and process 1 {6, 7, 8, 9, 10 }) and performs the partial sum of that array. In the end, each process uses MPI_Reduce to sum all the partial sums into a single value available to the master process (it could have been another process as well). Have a look at this example:
#include <mpi.h>
#include <stdio.h>
#include <stdlib.h>
int main(int argc, char* argv[]){
int np, pid;
MPI_Init(&argc, &argv);
MPI_Comm_rank(MPI_COMM_WORLD, &pid);
MPI_Comm_size(MPI_COMM_WORLD, &np);
int partial_sum = 0;
if (pid == 0) {
int a[] = { 1, 2, 3, 4, 5};
for(int i = 0; i < 5; i++)
partial_sum += a[i];
}
else if (pid == 1){
int a[] = {6, 7, 8, 9, 10};
for(int i = 0; i < 5; i++)
partial_sum += a[i];
}
int sum;
MPI_Reduce(&partial_sum, &sum, 1, MPI_INT, MPI_SUM, 0, MPI_COMM_WORLD);
if (pid == 0){
printf("Sum of array is : %d\n", sum);
}
MPI_Finalize();
return 0;
}
This code only works with 2 processes (and it is kind of silly( but I am using it to showcase the use of the MPI_Reduce.
I tried few solutions but they are not working, so can someone explain
few important changes to make to make it compatible with 2D array
also.
If you adapt your code to use the MPI_Reduce as I have shown, then it does not matter if it a 1D or 2D array, because you will first do the partial sum into a single value and then performance the reduction.
Alternatively, you can also have each row assigned to a process and then perform a reduction of the entire array, and then the master process performs the sum of the resulting array.
An example:
#include <mpi.h>
#include <stdio.h>
#include <stdlib.h>
int main(int argc, char* argv[]){
int np, pid;
MPI_Status status;
MPI_Init(&argc, &argv);
MPI_Comm_rank(MPI_COMM_WORLD, &pid);
MPI_Comm_size(MPI_COMM_WORLD, &np);
int partial_sum = 0;
int size = 5;
int a[5] = {1, 2, 3 , 4, 5};
int sum[5] = {0};
MPI_Reduce(&a, &sum, size, MPI_INT, MPI_SUM, 0, MPI_COMM_WORLD);
if (pid == 0){
int total_sum = 0;
for(int i = 0; i < size; i++)
total_sum += sum[i];
printf("Sum of array is : %d\n", total_sum);
}
MPI_Finalize();
return 0;
}
Output (for two processes):
Sum of array is : 30
This is the very basic function of my program, and as such is not necessarily reproducible. However, I was wondering if there is a way to send an array of arrays using MPI? Or is this something that is not possible and I should flatten my array? Any help would be greatly appreciated as I've been struggling with trying to figure this out.
int *individual_topIds;
int **cell_topIds;
cell_topIds = (int**) malloc(sizeof(int*)*25*boxes);
if(rank == 0) {
for (int i = 0; i < boxes; i++) {
individual_topIds = (int*) malloc(sizeof(int)*25);
for(int j = 0; j < cellMatrix[i].numTop; j++){
individual_topIds[j] = cellMatrix[i].aTopIds[j];
}
cell_topIds[i] = individual_topIds;
}
MPI_Send(cell_topIds, boxes*25, MPI_INT, 1, 10, MPI_COMM_WORLD);
}
Then in my rank == 1 section. I have tried send and receive with just boxes, and not boxes*25 as well.
for 1 -> boxes
MPI_Recv(cell_topIds, boxes*25, MPI_INT, 0, 10, MPI_COMM_WORLD, &status);
int *ptop;
ptop = (int*) malloc(sizeof(int)*25);
ptop = cell_topIds[i];
printf("1\n");
for(int j = 0; j < sizeof(&ptop)/sizeof(int); j++){
printf("%d, ", ptop[j]);
}
printf("2\n");
end for i -> boxes
free(ptop);
Edit: Forgot to mention that the output of the print is a seg fault
Caught error: Segmentation fault (signal 11)
This is not a particularly well-worded question.
However, MPI will let you send arrays of arrays if you use a custom type, as below:
#include "mpi.h"
#include <stdio.h>
struct Partstruct
{
char c;
double d[6];
char b[7];
};
int main(int argc, char *argv[])
{
struct Partstruct particle[1000];
int i, j, myrank;
MPI_Status status;
MPI_Datatype Particletype;
MPI_Datatype type[3] = { MPI_CHAR, MPI_DOUBLE, MPI_CHAR };
int blocklen[3] = { 1, 6, 7 };
MPI_Aint disp[3];
MPI_Init(&argc, &argv);
disp[0] = &particle[0].c - &particle[0];
disp[1] = &particle[0].d - &particle[0];
disp[2] = &particle[0].b - &particle[0];
MPI_Type_create_struct(3, blocklen, disp, type, &Particletype);
MPI_Type_commit(&Particletype);
MPI_Comm_rank(MPI_COMM_WORLD, &myrank);
if (myrank == 0)
{
MPI_Send(particle, 1000, Particletype, 1, 123, MPI_COMM_WORLD);
}
else if (myrank == 1)
{
MPI_Recv(particle, 1000, Particletype, 0, 123, MPI_COMM_WORLD, &status);
}
MPI_Finalize();
return 0;
}
Alternatively, use a flat array design (this is a good idea for performance reasons as well as easy compatibility with MPI).
Im trying to find a spesific value inside an array. Im trying to find it with parallel searching by mpi. When my code finds the value, it shows an error.
ERROR
Assertion failed in file src/mpid/ch3/src/ch3u_buffer.c at line 77: FALSE
memcpy argument memory ranges overlap, dst_=0x7ffece7eb590 src_=0x7ffece7eb590 len_=4
PROGRAM
const char *FILENAME = "input.txt";
const size_t ARRAY_SIZE = 640;
int main(int argc, char **argv)
{
int *array = malloc(sizeof(int) * ARRAY_SIZE);
int rank,size;
MPI_Status status;
MPI_Request request;
int done,myfound,inrange,nvalues;
int i,j,dummy;
/* Let the system do what it needs to start up MPI */
MPI_Init(&argc,&argv);
MPI_Comm_rank(MPI_COMM_WORLD,&rank);
MPI_Comm_size(MPI_COMM_WORLD,&size);
myfound=0;
if (rank == 0)
{
createFile();
array = readFile(FILENAME);
}
MPI_Bcast(array, ARRAY_SIZE, MPI_INT, 0, MPI_COMM_WORLD);
MPI_Irecv(&dummy, 1, MPI_INT, MPI_ANY_SOURCE, 1, MPI_COMM_WORLD, &request);
MPI_Test(&request, &done, &status);
nvalues = ARRAY_SIZE / size; //EACH PROCESS RUNS THAT MUCH NUMBER IN ARRAY
i = rank * nvalues; //OFFSET FOR EACH PROCESS INSIDE THE ARRAY
inrange = (i <= ((rank + 1) * nvalues - 1) && i >= rank * nvalues); //LIMIT OF THE OFFSET
while (!done && inrange)
{
if (array[i] == 17)
{
dummy = 1;
for (j = 0; j < size; j++)
{
MPI_Send(&dummy, 1, MPI_INT, j, 1, MPI_COMM_WORLD);
}
printf("P:%d found it at global index %d\n", rank, i);
myfound = 1;
}
printf("P:%d - %d - %d\n", rank, i, array[i]);
MPI_Test(&request, &done, &status);
++i;
inrange = (i <= ((rank + 1) * nvalues - 1) && i >= rank * nvalues);
}
if (!myfound)
{
printf("P:%d stopped at global index %d\n", rank, i - 1);
}
MPI_Finalize();
}
Error is somewhere in here because when i put an invalid number for example -5 into if condition, program runs smoothly.
dummy = 1;
for (j = 0; j < size; j++)
{
MPI_Send(&dummy, 1, MPI_INT, j, 1, MPI_COMM_WORLD);
}
printf("P:%d found it at global index %d\n", rank, i);
myfound = 1;
Thanks
Your program is invalid with respect to the MPI standard because you use the same buffer (&dummy) for both MPI_Irecv() and MPI_Send().
You can either use two distinct buffers (e.g. dummy_send and dummy_recv), or since you do not seem to care about the value of dummy, then use NULL as buffer and send/receive zero size messages.
If I want to define my own type, and use it as a datatype with MPI_Send to take only even rows from a matrix, does that matrix (send buffer) have to be allocated statically?
I seem to have problems when I allocate it dynamically. Is this because addresses need to be successive for data to be sent?
No, memory to be sent with MPI_Send does not have to be statically allocated.
To send array subsets, you likely want to use MPI_Type_indexed. Here is a slightly modified version of the example from the mpi.deino.net article on MPI_Type_indexed, where I have replaced the statically allocated buffer
int buffer[27];
to a dynamically allocated buffer
int* buffer = (int*)malloc(27 * sizeof(int));
I hope it helps:
#include <mpi.h>
#include <stdio.h>
int main(int argc, char *argv[])
{
int rank, size, i;
MPI_Datatype type, type2;
int blocklen[3] = { 2, 3, 1 };
int displacement[3] = { 0, 3, 8 };
int* buffer = (int*)malloc(27 * sizeof(int)); //int buffer[27];
MPI_Status status;
MPI_Init(&argc, &argv);
MPI_Comm_size(MPI_COMM_WORLD, &size);
if (size < 2)
{
printf("Please run with 2 processes.\n");
MPI_Finalize();
return 1;
}
MPI_Comm_rank(MPI_COMM_WORLD, &rank);
MPI_Type_contiguous(3, MPI_INT, &type2);
MPI_Type_commit(&type2);
MPI_Type_indexed(3, blocklen, displacement, type2, &type);
MPI_Type_commit(&type);
if (rank == 0)
{
for (i=0; i<27; i++)
buffer[i] = i;
MPI_Send(buffer, 1, type, 1, 123, MPI_COMM_WORLD);
}
if (rank == 1)
{
for (i=0; i<27; i++)
buffer[i] = -1;
MPI_Recv(buffer, 1, type, 0, 123, MPI_COMM_WORLD, &status);
for (i=0; i<27; i++)
printf("buffer[%d] = %d\n", i, buffer[i]);
fflush(stdout);
}
MPI_Finalize();
free(buffer);
return 0;
}
I was wondering why this program actually works in MPI (openMPI 1.5/1.6. )
#include <stdio.h>
#include <mpi.h>
#define VECTOR_SIZE 100
int main(int argc,char ** argv) {
int A[VECTOR_SIZE];
int sub_size=2;
int count=10;
MPI_Datatype partial_array;
int rank,size;
MPI_Status status;
MPI_Init(&argc,&argv);
MPI_Comm_rank(MPI_COMM_WORLD,&rank);
MPI_Comm_size(MPI_COMM_WORLD,&size);
MPI_Type_vector(count, sub_size,
2*sub_size, MPI_INT, &partial_array);
MPI_Type_commit(&partial_array);
if (rank == 0) {
int i;
// server - initialize data and send
for (i = 0; i< VECTOR_SIZE; i++) {
A[i] = i;
}
MPI_Send(&(A[0]), 1, partial_array, 1, 0, MPI_COMM_WORLD);
} else if (rank==1) {
int i;
for (i = 0; i< VECTOR_SIZE; i++) {
A[i] = 0;
}
// vector is composed by 20 MPI_INT elements
MPI_Recv(&(A[0]),20, MPI_INT, 0, 0, MPI_COMM_WORLD, &status);
printf("\n");
for (i = 0; i<VECTOR_SIZE; i++) {
printf("%d ",A[i]);
}
printf("\n");
}
MPI_Finalize();
}
while this other program where Send and Receive primitives are exchanged does not terminate (the receive never completes):
#include <stdio.h>
#include <mpi.h>
#define VECTOR_SIZE 100
int main(int argc,char ** argv) {
int A[VECTOR_SIZE];
int sub_size=2;
int count=10;
MPI_Datatype partial_array;
int rank,size;
MPI_Status status;
MPI_Init(&argc,&argv);
MPI_Comm_rank(MPI_COMM_WORLD,&rank);
MPI_Comm_size(MPI_COMM_WORLD,&size);
MPI_Type_vector(count, sub_size,
2*sub_size, MPI_INT, &partial_array);
MPI_Type_commit(&partial_array);
if (rank == 0) {
int i;
// server - initialize data and send
for (i = 0; i< VECTOR_SIZE; i++) {
A[i] = i;
}
MPI_Send(&(A[0]),20, MPI_INT, 0, 0, MPI_COMM_WORLD);
} else if (rank==1) {
int i;
// client - receive data and print
for (i = 0; i< VECTOR_SIZE; i++) {
A[i] = 0;
}
MPI_Recv(&(A[0]), 1, partial_array, 1, 0, MPI_COMM_WORLD, &status);
printf("\n");
for (i = 0; i<VECTOR_SIZE; i++) {
printf("%d ",A[i]);
}
printf("\n");
}
MPI_Finalize();
}
If I understand MPI type mathing rules correctly neither of the two should complete.
Obviously in the second program rank 0 is sending to itself and rank 1 is expecting message also from itself:
MPI_Send(&(A[0]),20, MPI_INT, 0, 0, MPI_COMM_WORLD);
destination rank should be 1, not 0
MPI_Recv(&(A[0]), 1, partial_array, 1, 0, MPI_COMM_WORLD, &status);
source rank should be 0, not 1.
Otherwise you do not understand the MPI type matching correctly. It only states that underlying primitive types in the type maps on both ends should match. You are creating a vector whose type map has 20 primitive integers. If you send one element of this type, your message will actually contain 20 integers. On the receiver side you provide space for at least 20 integers so this is correct. The opposite is also correct.
It is not correct if you send only 10 or 18 integers in the second program since they will not make a complete element of the vector type. Nevertheless, the receive operation will complete but if you call MPI_Get_count() on the status, if will return MPI_UNDEFINED because from the number of received primitive integer elements one cannot construct an integer number of vector elements. It is also not correct to mix primitive types, e.g. send MPI_DOUBLE (or vector, or structure, or whatever other type that has doubles) and receive it as MPI_INT.
Please also note that MPI messages do not carry their type map or type ID with them so most MPI implementations do not check if types match. It is possible to send MPI_FLOAT and receive it as MPI_INT (because both are 4 bytes on most systems) but it is not correct to do so.