Motivated by the limit on the power usage effectiveness (PUE) of the data centers, the potential benefit of the consolidation, and the impetus of achieving maximum return on investment (ROI) on the cloud computing market, we investigate VM placement in the data center, formulate a multi-level generalized assignment problem (MGAP) for maximizing the profit under the service level agreement and the power budget constraint based on the model of a virtualized data center, and solve it with a first-fit heuristic. Numerical simulations show that the first-fit heuristic is effective in solving the large-scale instances of the MGAP with the sampled simulation setups.