Question regarding using MOSEK in CVX

Hello, this question may seem stupid, but I got confused.

I plan to solve a SDP problem with a matrix of size more than 1500*1500. I’m planning to use MOSEK as a solver, since MOSEK is somehow designed to solve large-scale SDP problems. But I found the following sentence in CVX documentation: “CVX is not meant for very large problems, so if your problem is very large (for example, a large image processing or machine learning problem), CVX is unlikely to work well (or at all). For such problems you will likely need to directly call a solver, or to develop your own methods, to get the efficiency you need.”

These sentences confused me. Does this mean by using MOSEK as a solver in CVX, I can only partly use MOSEK’s computational power? What is the difference between directly call MOSEK outside and set MOSEK as a CVX solver?

It just means that CVX can take a lot of time to process your model before it calls MOSEK. Depending on the structure and size of models this step can be very time and memory consuming.

Got it, thanks for the answer