I made a spreadsheet to plot the 3d speed distribution from the MB probability function. It matches the peak and fall-off of published graphs. I then tried to integrate it by summing over interval widths times probability. I thought the area should be 1. My result is around 5 or 6. I've tried really small intervals, around 0.25 m/s, that didn't help. Must be I don't understand what the area under the probability curve represents.
Check out this article: http://en.wikipedia.org/wiki/Maxwell–Boltzmann_distribution It gives properties of the M-B distribution, including the PDF and corresponding CDF. You must have made an error in your calculations.
You have failed to normalize! For example, see http://hyperphysics.phy-astr.gsu.edu/hbase/kinetic/maxspe.html