In the stress simulation, such as static, we always use symmetry to simplify the simulation, when the part is symmetrical and the condition is symmetrical too. However, I always doubt whether the symmetry also works in flow simulation.
For example, there is a high-speed flow in a pipe. The turbulence always make it complicated, and not symmetrical in the real test. Should we use symmetry only in low-speed and laminar flow?
This is a good question, unfortunately there is no easy answer. For instance, the alternating vortex shedding that can be observed behind blunt bodies in steady flow can happen even at low Re flow. This is one of those things where you will have to draw on engineering experience.
One thing you could do is do a full domain simulation with a coarse mesh and then study those results to see if a finer mesh half- or quarter-domain study will be an appropriate model.