We study decentralized derivative-dependent control of large-scale second-order systems with input delay via delayed feedback implementation. The derivatives are approximated by finite differences leading to a time-delayed feedback. In the centralized case, an efficient simple linear matrix inequities (LMIs)-based method for designing of such static output-feedback and its sampled-data implementation was recently suggested. In the present paper, we extend this design to large-scale systems in the presence of input delay. We add appropriate terms to the corresponding Lyapunov-Krasovskii functional to compensate the terms due to the input delay. Numerical example illustrates the efficiency of the method.