Question

If I run this test:

 var r = new Random();
 var ints = new int[13];
 Parallel.For(0, 2000000, i => {            
     var result = r.Next(1, 7) + r.Next(1, 7);
     ints[result] += 1;
 });

I get the following result:

2: 92,14445
3: 0,41765
4: 0,62245
5: 0,82525
6: 1,04035
7: 1,25215
8: 1,0531
9: 0,8341
10: 0,6334
11: 0,4192
12: 0,2109

When I use the regular For:

for (int i = 0; i < 2000000; i++) {
    var result = r.Next(1, 7) + r.Next(1, 7);
    ints[result] += 1;
}

The output is:

2: 2,7797
3: 5,58645
4: 8,3414
5: 11,09935
6: 13,8909
7: 16,6731
8: 13,82895
9: 11,10205
10: 8,3424
11: 5,5712
12: 2,7845

The last result is a Triangular Distribution and it is the expected output.

The purpose of my question is not to discuss the applicability of parallelism. The question is why the Parallel.For behaves that way?

Answer 1

The Random class methods are not thread safe.

http://msdn.microsoft.com/en-us/library/system.random.next(v=vs.90).aspx#2

So the first piece of code is just demonstrating some undefined behavior.

EDIT:

As for a little speculation, from what little I know about operating systems I believe random number generation is a pretty low level operation and hence might even require a context switch. While this is happening you may end up grabbing the same random number multiple times before it’s had a chance to update. This would account for the lopsided distribution.