We compare the discovery potential of the LHC for lepton number violating (LNV) signals with the sensitivity of current and future double beta decay experiments, assuming 0 nu beta beta decay is dominated by heavy particle exchange. We consider charged scalar, leptoquark and diquark mechanisms of 0 nu beta beta decay, covering the 0 nu beta beta decay operators with both, the smallest and largest, possible rates. We demonstrate, if 0 nu beta beta decay were found with a half-life below 10(26)-10(27) years a positive signal should show up at the LHC, except for some particular cases of the leptoquark mechanism, and vice versa, if the LHC does not find any hints for LNV, a ¿short-range¿ explanation for a finite 0 nu beta beta decay half-life will be ruled out in most cases. We argue, if a positive LNV signal were found at the LHC, it is possible to identify the dominant contribution to 0 nu beta beta. Two different kinds of observables which could provide such ¿model discriminating¿ power are discussed: different invariant mass peaks and the charge asymmetry.