Python is used to glue all the lower level stuff together. If you are iterating over large arrays in python you are doing something wrong, but if you are using python to glue C/Fortran molecules together then it starts to look much better.
tpjunkie writes: Many slashdot readers will remember D-wave's announcement in 2007 of its quantum computer, an announcement met with skepticism and a good amount of scorn. However, today the company has sold quantum computers to such companies as Lockheed Martin and Google, and their computers have gone from a handful of qubits to 512 in their most recent offerings. Nature has a story including an interview with the company's founder Geordi Rose, and a look at where the company is headed and some of the difficulties it has overcome.
Our aircraft carriers are longer then yours!
On a more serious note the largest calculation that I can find (in terms of # of cores utilized) was a fluid dynamics calculation with a million cores on Sequoia. From my own experience we usually utilize 4-100 cores for throughput over the speed of a single job- if it takes a month to do then so be it.