This paper presents a strategy for controlling generators in a small-footprint power system without the need for a centralized decision maker. In particular, we discuss an iterative algorithm which is utilized to coordinate a set of generators such that they collectively operate according to some predetermined criterion while accounting for individual capacity constraints. We implement the algorithm using a hardware testbed comprised of low-complexity devices capable of performing simple computations and communicating wirelessly with other nearby devices. The hardware testbed is then used to control the generation output of the synchronous generators in a six-bus, small-footprint power system. Experimental results are presented for several load changes as well as for a case in which a spinning reserve is added to the system.