Steven> I'm not convinced that the average is meaningful.
It's not, that's why I included the standard deviation and min/max as well.
boost::fast_pool_allocator is not intended to be used with std::vector. You're averaging many cases for which it is documented to behave badly with a couple of cases for which it is fine.
That's correct. I was asked to compare with boost::pool_allocator, boost::fast_pool_allocator, std::allocator and tbb::allocator.
Also, even though pool_allocator is supposed to work with std::vector, it is slow as I would expect. The pool data structure is really designed for fixed size allocations and I for one am not particularly enamored of the idea of using it for std::vector.
Also, this is completely unreleated, but for lines like this
fast 1.49 0.838 0.603 1.32e+003
It looks like the cumulative min/max are being used instead of the local min/max.
That's also correct and I noticed this as well. I'll change it so that the test summaries show min/max just for that test, and keep the overall the global min/max.
In Christ, Steven Watanabe
Regards, Christian.