That's less than 2.5%.

When you consider that any optimisation is likely to provide better returns, then this really isn't a priority. The techs want to make things better, but (and I think this has been said before) they want to go after the low-hanging fruit - the things that provide the best return for their time.

Reducing the pagefault problem is likely to be very challenging. If we saw the 20+% being lost that we saw with an earlier project, then the techs would make this a high priority. But 2.5%? Not a chance.