Hard processes in proton-proton (pp) collisions, composed by partonic interactions with momentum transfer larger than the QCD scale, might occur together with lower transverse momentum processes, called Multiple Parton Interactions (MPI). MPI are as ubiquitous in pp collisions at LHC, as multiple nucleon interactions in heavy-ion collisions. In this talk, we will discuss similarities between the two concepts and how they manifest themselves in observables.
It was recently shown  that multiplicity as a function of the number of participant nucleons in heavy-ion collisions rises steeply for very central events. The most far reaching interpretation of this result is breaking of participant scaling. We will discuss how this result can be understood in terms of measurement biases, not taking into account the fact that a nucleus can run out of participating nucleons. More surprising is the seemingly similar result from pp collisions with a high momentum transfer process . Here an analogous interpretation – at least in the case of standard MPI-based models, such as the Pythia one – is not as straight forward.
Keeping these features in mind, some possibilities of improving MPI modelling using a recent implementation of the Mueller dipole model  will be presented.
 ALICE collaboration, Phys. Lett. B790 (2019), 35-48, arXiv:1805.04432 [nucl-ex]
 ALICE collaboration, Eur. Phys. J. C79 (2019) no.10, 857, arXiv:1905.07208 [nucl-ex]
 C. Bierlich and C.O. Rasmussen, JHEP 1910 (2019) 026, arXiv:1907.12871 [hep-ph]