During the presentation, we will discuss some selected aspects of the simulation of wall-bounded flows, both focusing on the question of stability and transition to turbulence, but also fully turbulent flows. In a first part, we study the case of a jet in crossflow, which is a common occurrence for instance in combustion engines. Using Nek5000, we look at method to describe the first bifurcation of the flow. It turns out that this global stability problem, solved by means of an Arnoldi method, is very sensitive and thus maybe not as relevant for practical predictions. Therefore we compute, for the first time in this context, optimal initial conditions/disturbances, which are the likely way for transition to occur. In a second part, we will focus on fully turbulent flows. We start with the example of spatially developing boundary layers, but consider mainly with turbulent pipe flow in straight and bent pipes. Some interesting and counter-intuitive findings such as sub-laminar and sub-straight drag will be discussed. Another aspect which has recently gained considerable attention, is the study of coherent vortical structures in wall turbulence. Using our data sets we will try to identify which structures are indeed observed in near-wall turbulence. In particular, we will show that the so-called hairpin vortex is a frequent occurrence close to transition, but unlikely to be observed at higher Reynolds numbers.