Although nowadays multi core systems are a so-to-say standard it is not self-evident for applications to make use of all available cores. This is because not all tasks can be parallelized.
Not so in the area of software development with Visual Studio and C/C++, where Microsoft provides a possibility to compile multiple projects within one solution in parallel. Sadly the performance of this approach strongly depends on the architecture of the solution, meaning the number and size of the projects and their dependencies. In addition some single projects typically are being compiled more often than the whole system, compromising the integrated parallelization approach.

Wouldn't it be nice to always utilize all available cores for compilation, apart from the solutions' dependencies? The solution is Shark Compiler Control.