Performance Issues with loops in JavaScript

Question

I'm having huge performance problems with my JavaScript code. I try to pre-process data for a Cplex-Model in OPL, the code is written in JavaScript.

I already tried to speed it up by adding the first if-clause (if A[1][3]="", A[1][k] ="" for every k > 3) , but it still takes my computer more than 2 hours to process just this code...

Jnr = 1000,
Mnr = 35;

while (j <= Jnr) {
  while (i <= Mnr) {
    while (k <= Mnr) {
      if (A[j][k] == "") {
        break;
      } else if (A[j][k] == I[i].name) {
        counter[j][i] = counter[j][i] + 1;
      }
      k = k + 1;
    }
    k = 1;
    i = i + 1;
  }
  i = 1;
  j = j + 1;
}
j = 1;

Is there anyway to speed up the process?

I appreciate your help

Answer 1

It's not a very good idea to use scripting for big loops. If you write that directly in OPL this will be much faster:

int Jnr = 1000;
int Mnr = 35;

tuple t
{
string name;
}

t I[i in 1..Mnr]=<"A">;

string A[i in 1..Jnr][j in 1..Mnr]=((i+j)%2==0)?"A":"B";
int counter[1..Jnr][1..Mnr];

execute
{
var i=1;
var j=1;
var k=1;

while (j <= Jnr) {
  while (i <= Mnr) {
    while (k <= Mnr) {
      if (A[j][k] == "") {
        break;
      } else if (A[j][k] == I[i].name) {
        counter[j][i] = counter[j][i] + 1;
      }
      k = k + 1;
    }
    k = 1;
    i = i + 1;
  }
  i = 1;
  j = j + 1;
}
j = 1;
}


int counter2[i in 1..Jnr][j in 1..Mnr]=sum(k in 1..Mnr) (I[k].name==A[i][k]);

execute
{
counter2;
}


assert forall(i in 1..Jnr,j in 1..Mnr) counter[i][j]==counter2[i][j];

counter takes 9 s on my machine, whereas counter2 takes 0.3 s