We have carried out Monte Carlo (MC) simulations using the embedded atom potential to study the sliding and migration of the σ5  (210) tilt grain boundary (GB) in aluminum and the effect of vacancies on the sliding properties. We find that the simulated annealing allows the system to gradually anneal to a global-minimum configuration, thus increasing the number of migrations and reducing the GB sliding energy barriers to about a factor of three compared to the corresponding “static” values. The distribution of atomic energies as a function of GB displacement, provide insight into which atoms are responsible for the GB migration. The vacancy formation energy is found to be lower when the vacancy is placed on the first layer to the boundary, in excellent agreement with ab initio electronic structure calculations. The sliding and migration properties depend very sensitively on the position of the vacancy in the GB core.