Variability is the key concept to facilitate multivariance in computer applications. Static variability is
applied at compile-time and allows developers to decide which parts of their software should be part
of the final executable. Its benefit is a tailored binary file which only includes the data necessary
to perform its tasks while omitting other data parts. A drawback is that not every decision can be
made at compile-time which is why dynamic variability allows to defer that decision to the runtime
of the application. During execution, it alters its behavior based upon its internal state to provide
the desired variability. While this is a benefit, it also increases the runtime latency as the program
has to evaluate its internal state each and every time. Even if the internal state has not changed.
This thesis bridges the gap between static and dynamic variability and tries to combine the
benefits of both approaches while reducing their individual drawbacks. During this thesis a linker
has been extended to allow building and embedding tailored application views inside the executable.
With the help of a runtime library the threads can individually switch to these specialized views to
benefit from static variability. The library creates dedicated Address-Space Views which contain the
views to which the application threads can dynamically switch to during execution, thus combining
static with dynamic variability. The evaluation shows that the established method can be applied to
real-world applications. Additionally, it can provide a significant positive performance impact in
certain scenarios. Individual tailored views also allow to harden security in multithreaded computer
applications by locally restricting the code a thread can execute.