Sparse arrays vs full arrays

JavaScript performance comparison

Test case created by Mattias Ernelli

Preparation code

 
<script>
Benchmark.prototype.setup = function() {
    var n, N, i, I, sparse = [],
        full = [],
        full2 = [];
   
    N = 16384;
    I = 8;
   
    for (i = 0; i < I; i++) {
      do {
        n = (N * Math.random()) | 0;
        if (n > (N - 1)) {
          n = N - 1;
        }
      } while (sparse[n]);
      sparse[n] = -100 + 200 * Math.random();
    }
    full.length = N;
    full2.length = N;
    for (i = 0; i < N; i++) {
      if (typeof sparse[i] === "number") {
        full[i] = sparse[i];
        full2[i] = sparse[i];
      } else {
        full[i] = 0;
        full2[i] = -100 + 200 * Math.random();
      }
    }
};
</script>

Test runner

Warning! For accurate results, please disable Firebug before running the tests. (Why?)

Java applet disabled.

Testing in unknown unknown
Test Ops/sec
Sparse array sum
var sum = 0;
sparse.forEach(function(v, i) {
  sum += v;
});
pending…
Full array sum
var len = full.length;
sum = 0;
for (i = 0; i < len; i++) {
  sum += full[i];
}
pending…
Running sparse on full array
var sum = 0;
full.forEach(function(v, i) {
  sum += v;
});
pending…
Sum all elements on a full array without zeros
var len = full2.length;
sum = 0;
for (i = 0; i < len; i++) {
  sum += full2[i];
}
pending…

Compare results of other browsers

Revisions

You can edit these tests or add even more tests to this page by appending /edit to the URL. Here’s a list of current revisions for this page:

1 comment

Mattias Ernelli (revision owner) commented :

OK, thats it, Sparse arrays in JS is way slower than just processing all elements of the array. Even if less than 1% of the sparse array is populated.

I will remove all code supporting sparse arrays in my numjs library which I'm working on from time to time...

Add a comment