To be honest, I can't actually tell you how to use the function because it's not documented in DGDK either. The second argument requires an address to an array in memory, but not a managed array. So for example, it could be potentially used in C++ if you've created a correctly formatted array structure in memory and you pass the address of the array.
I believe that the function was used as part of the DBP language for managing internal array creation of sorts, but unless Mike or Lee explains what the function is for, I can't say any more on it.
It's only available in the .NET toolkit because it was also available in DGDK (C++). Both toolkits don't explain the function's use in their respective documentation.
Paul.