This talk illustrates the use of optimization based methods to provide insight into the analysis, design, and operation of a power system with large-scale storage integration. We first discuss optimal power flow (OPF) based models for storage dispatch and allocation throughout the network. We then exploit the structure of the optimization problem to form a storage subproblem that isolates the relationship between the storage variables and market based price signals. The storage subproblem is used to show that these price signals are critical factors in optimal storage dispatch and siting decisions. The second part of the talk seeks to reduce the computationally complexity of the OPF+storage problem by proposing a simplified model that augments commonly used linearizations through the inclusion of a quadradic loss approximation. This so-called DC optimal power flow (DCOPF) with losses formulation is then used to analyze the role of real power losses in optimizing grid-scale storage integration. We demonstrate the accuracy and improved computationally tractability of this method through case studies using a wind integrated test system.