CMake and C++ for Scientific Computing

C++ is a great language for scientific computing. For most people, however, getting started with C++ is a big hurdle. Most people think that it is hard to setup a C++ project and write correct/nice code. With the use of CMake and the Eigen and Boost libraries however, developing a C++ project for a scientific computing task will be almost as easy as writing a Python project with NumPy.

The only thing that is a bit more involved is the initial setup. On the following two pages I will show you how to get started with a CMake project on both windows and Linux.

The libraries we will use are

• Boost for general functions, like string formatting and file system access
• Eigen it is a library for linear algebra, but can be thought of as a NumPy equivalent for C++.
• OpenMP a very friendly API for parallel programming