Most of these questions don’t have answers as of now, but we can make some remarks:

- It is believed by arguments by t’Hooft, that large N gauge theories have dual descriptions as string theories. Only a few examples are known, the most famous being $\mathcal{N}=4$ SYM, whose string theory dual lives on 5d anti-de-Sitter space. One of the hopes is to find a similar description for QCD.
- The non-abelian refers to the non-abelian gauge group. Roughly the flux tubes connecting quarks are string-like. This doesn’t have much to do with the statistics of the strings (unless you introduce magnetic monopoles).
- Yes, the string should have a worldsheet description. The problem is that the spacetime on which the string moves is not the same as the spacetime on which the gauge theory lives. This is the hard part of the problem. The string should be described by a non-linear sigma model on some target space. There is currently no precise mathematical procedure for extracting this information from a gauge theory such as QCD.

Bonus: like one of the comments said, the supersymmetric case is actually simpler, since we can identify dual descriptions more easily and compute things to check these dualities.

This post imported from StackExchange Physics at 2020-10-28 19:07 (UTC), posted by SE-user Adolfo Holguin