University of Cologne
Julia has the potential to become the major programming language in numerical physics. In this presentation I will share my personal story about how switching to Julia has improved my physics research as well as outline some issues I have encountered. I will explain how I replaced an opaque C++ quantum Monte Carlo code by a clean and simple Julia implementation. While showing similar performance in large scale simulations of so-called metallic quantum critical points the Julia code has an order of magnitude fewer lines and allows one to perform number crunching and analysis in the same framework. I will argue that Julia’s type system and multiple dispatch paradigm naturally leads to flexible code which can be used in many different projects. As an implementation of this idea, I will conclude by introducing MonteCarlo.jl, a new package for simulating physical systems such as spin models and itinerant fermions by various (quantum) Monte Carlo flavors.
Carsten is a Ph.D. student in the theoretical physics department of the University of Cologne working on quantum criticality in metals. He studied physics and computer science at the Goethe University Frankfurt, the University of Florida, and the TU Darmstadt.
