Yes, I believe we are seeing the same behavior.  I'm not sure this is a bug
in one sense, since the program ultimately produces the correct output
(an "out of memory" exception).  It seems that the runtime tries really hard
to keep going when there isn't much memory, which may be the right thing to do.
But from a user point of view, it is rather disconcerting that program
execution slows down so much when almost out of available memory.  Maybe there
is no better solution than the current implementation.  Or perhaps there could
be some way to tune how hard the memory allocation routines try to coalesce
chunks when low on memory.