Number of gravitons launched by a proton The wikipedia article http://en.wikipedia.org/wiki/Gauge_bosons describes how in QM exchanges of gauge bosons carry force, and describes how the graviton may also be a gauge boson.
If the observable universe is made of around 10^80 hydrogen atoms*, this implies each atom should be exchanging gauge bosons with 10^80 other atoms. So each atom should launch 10^80 gravitons, and a similar number of photons, gluons etc.
Is this really how it works?  Does one proton launch 10^80 bosons when it comes into existance?
* http://www.universetoday.com/36302/atoms-in-the-universe/
 A: It is meaningless to talk about the precise number of virtual particles corresponding to the interaction of a certain particle with other particles. It is simply not well-defined, neither in the mathematical framework of the path integral formalism of quantum field theory, nor if one uses them as a heuristic picture of reality. Virtual particles appear in a context where one needs properly defined in- and out-states, then one can formulate a path integral and draw Feynman diagrams, and in principle there are infinitely many of them. As virtual particles correspond to internal lines in Feynman diagrams, this circumstance makes counting them quite hard. Heuristically, one could say that we only count those diagrams that contribute the most to a given process and say that it is mediated by the corresponding virtual particles. However, one cannot formulate the process of the interaction of a proton with "all other atoms" in such a way that is is possible to acquire a meaningful answer.   
A: Since the commenters are not turning their comments to an answer, I will summarize:

If the observable universe is made of around 10^80 hydrogen atoms*, this implies each atom should be exchanging gauge bosons with 10^80 other atoms.

So far so good. The field theoretical view of interactions says that these happen by the exchange of virtual particles between "real particles", real means that they are on their mass shell.
This Feynman diagram 

Shows the on shell particles: The kaon which decays to three pions,  which have definite mass, and off shell ones,  the quarks and the gluon and the W which are off mass shell and virtual and thus do not obey energy and momentum conservation laws. This is most striking in the case of the W exchange, where the mass of the W on shell (real) is orders of magnitude larger than the mass of the real Kaon. They are just quantum number representations similar to the correspondingly named  on shell particles ( gluons and quarks are never free, so never on shell, but that is another story).

So each atom should launch 10^80 gravitons, and a similar number of photons, gluons etc.t

No. Each atom theoretically can interact with every other atom through virtual particle exchanges, though these interactions will be extremely extremely weak the larger the distance between the on shell particles. These off shell particles may be one or many, from each hydrogen atom, though in the calculation usually a few exchanges are dominant.
Launch implies on shell manifestation, in my vocabulary. 
What you should keep is that there is no conservation on the number of off the mass shell virtual particles other than the diminishing probability of multiple exchanges, diminishing  at least for weak, electromagnetic and gravitational interaction. Gluons are another type of animal and another story too.
