cewit's newsletter ran a short piece on what i was doing with the university's supercomputing allocation: audio synthesis and modelling at a scale that does not fit on a desktop.
the useful thing about having that much compute is that it changes which questions you are allowed to ask. a lot of audio engineering is choosing a filter design or a basis function on the strength of experience and then tuning it until it sounds right. with a cluster you can search the space instead — evaluate thousands of candidates against a real corpus and find out whether the conventional choice was ever the best one.
the same models turn up well outside music. forensic audio and medical imaging are both, underneath, the problem of pulling a signal out of noise when you have a decent prior about what the signal should look like.
read the full article — it is inside the february 2016 newsletter pdf.