The absence of common image storage, processing and visualization approaches is a very serious problem of the collaboration between researchers working in different image cytometry laboratories. The results produced by individual groups working in the same field are therefore often not comparable. Even a simple visualization of image data acquired in a different laboratory is often time-consuming because of the non-existence of a common 3D image file format. Recently we presented a solution to these problems for the specific task of acquisition, storage, processing and visualization of 3D image data of fluorescently stained genetic material used in the research of the spatial organization of human genome [7]. We have developed a special software system based on the client/server architecture that enables the collaboration of laboratories active in this field. Each laboratory acquires the 3D image data using its own microscopes but stores it together with a detailed description of samples on a central high-performance multi-processor UNIX server that acts as a file server, database server and image analysis server that accepts orders from individual clients. The requested parts of image data as well as image analysis results are sent back to the clients via Internet. In order to provide high computational power the server was equipped with the fastest processors available and the whole network architecture has been optimized for speed. Although the system has primarily been developed for human genome studies its architecture is well applicable also in other research fields. In this contribution we will present the main idea how to effectively share microscopy and image analysis resources as well as what possibilities the networked solution offers.