Groups and clusters of galaxies occupy a special position in the hierarchy of large-scale cosmic structures because they are the largest and the most massive (from ~10^13 Msun to over 10^15 Msun) objects in the universe that have had time to undergo gravitational collapse. The large masses of clusters imply that their contents have been accreted from regions of ~8-40 comoving Mpc in size and should thus be representative of the mean matter content of the universe. During the next decade sensitive multi-wavelength observations should be able to map the radial distributions of all main mass components (stars, cold, warm, and hot gas and total mass) at z<~ 1 out to the virial radius. At the same time, comparative studies of real and simulated cluster samples sould allow us to use clusters as veritable astrophysical laboratories for studying galaxy formation, as well as testing our theoretical models of structure formation and underlying assumptions about fundamental physics governing the universe.