The paper discusses the prospect of implementing a software package for the optimal scheduling of large-scale water supply and distribution networks. The time distribution function methodology is considered to be a suitable candidate for the solver implementation. However, since the existing methodology enables optimization of middle size networks, a concept of spatial decomposition is introduced and investigated. The resulting optimization problems are convex and soluble by the existing numerical methods of mathematical programming. Effective measures are proposed to reduce computational complexity of the approach. Finally software technologies to be used for the implementation are presented along with a relational data model for the problem.