The smart grid integrates advanced sensors, a two-way communication infrastructure, and high-performance computation-based control. The distribution management systems for smart grid include several functions for manipulating legacy voltage control devices and distributed energy resources through closed-loop volt/var control, leading to wide-area regulation of voltages in the presence of fluctuating power. The other primary distribution network analysis application is concerned with automatic fault location and service restoration following fault events, aiming to provide the grid with autonomous intelligence for self-healing. Communication technologies are vital to enable the computing applications of distribution networks, whether they work in centralized or distributed modes. This paper presents the state of the art in distribution management system architectures and modern workflows showing data exchange, practical parallel implementations designed to handle large amounts of data, in addition to communication standards that serve as interoperability enablers. It demystifies the relationship between different functions developed independently by power system researchers and shows their operation as a complete system, thus placing them in a better context for future research and development.