MapCloud is a solution combing four different Software Oriented Architecture (SOA) methods to obtain a reliable and efficient delivery method for digital map data, more precisely 3D map data.Mapping software use large amount of data including mapping data, aerial imagery and 3D models. Pre-processing data is a cumbersome task due the sheer amount of data that needs to be handled. The challenge is also on the other end, when the data is accessed in real-time. This process needs to be fast and reliable. This theses is done at Agency9. Agency9 is a Swedish company offering mapping and Geographic Information System (GIS) software and services worldwide.Agency9’s current system and workflow is limited to a set of single high performing servers for processing and preparation. In the workflow there are also manual steps that constitute a bottleneck for administration due to the need of copying huge amounts of data. From a design standpoint MapCloud overcomes these shortcomings. The workflow of MapCloud utilizes parallel data processing for preparation and distributed, resilient, access method with high-throughput for web-centric map applications.