The reliable performance of the smart grid is a function of the configuration and cyber-physical nature of its constituting sub-systems. Therefore, the ability to capture the interactions between its cyber and physical domains is necessary to understand the effect that each one has on the other. As such, the work in this paper presents a co-simulation platform that formalizes the understanding of cyber information flow and the dynamic behavior of physical systems, and captures the interactions between them in smart grid applications. Power system simulation software packages, embedded microcontrollers, and a real communication infrastructure are combined together to provide a cohesive smart grid cyber-physical platform. A data-centric communication scheme, with automatic network discovery, was selected to provide an interoperability layer between multi-vendor devices and software packages, and to bridge different protocols. The effectiveness of the proposed framework was verified in three case studies: (1) hierarchical control of electric vehicles charging in microgrids, (2) International Electrotechnical Committee (IEC) 61850 protocol emulation for protection of active distribution networks, and (3) resiliency enhancement against fake data injection attacks. The results showed that the co-simulation platform provided a high-fidelity design, analysis, and testing environment for cyber information flow and their effect on the physical operation of the smart grid, as they were experimentally verified, down to the packet, over a real communication network.2 of 17 researchers developed a co-simulation framework through which they combined OMNeT++ and OpenDSS for communication networks and power system simulations to examine wide area monitoring applications. Also, researchers in [4] showed a framework to simulate a power-routing algorithm for microgrid-clusters by combining OMNeT++ with real-time digital simulators. In [5], the authors pointed out the existence of a huge research gap regarding the simulation of cyber-physical systems and recommended that more research is required in this area. Therefore, they presented an event-driven co-simulation scheme utilizing Network Simulator NS2 and OpenDSS. In [6], a co-simulation setup based on IEC 61850 was presented for a low-voltage grid. MATLAB toolboxes SimEvents and Sim Power Systems were utilized for the modeling of the information flow between the system's cyber and physical layers, respectively. The aforementioned research denotes a significant step to achieve proper modeling techniques for the physical and cyber domains of cyber-physical systems. However, these schemes are not being deployed over an actual communication network. Accordingly, capturing the practical issues associated with high fidelity will be difficult since they are limited to the network simulation software's provided functionalities. For example, network simulators usually model networks on the large-scale using probabilistic and statistical models to forecast delays; they do not function on th...