The InfoGlue architecture was designed to benefit as much as possible from other OSS-projects as possible. We have throughout the project made use of other Open Source modules to be able to focus on the business logic of the platform instead of on support functionality like database persistence, image handling, application flow and visualization techniques. The result is a very high level of stability, modularity and speed of development.

As stated we feel we could not have managed to produce such a complex platform this fast without the other projects. Below is a simplified schetch over the application layers and how the framworks fit in.

Deployment environments

Deployment of the platform is currently supported on all platforms which can run Java. To run InfoGlue you don't need a full fledged EJB-server but a just a common web-container like Apache Tomcat is enough and even recommended. We know of people running it on IBM Websphere and Bea Weblogic as well but they are optional. The same principle applies to the database environment. You can run InfoGlue on MySQL which is free but if you want you can also use DB2 or Oracle or SQLServer for example.

Below is a very rough schetch of how a medium/small deployment could look like. The proposed setup can be made entirely without license fees for anything if you choose to run InfoGlue on Linux and MySQL for example.

The deployment is of course scalable so you can have multiple servers running both tools and websites. You are also able to have multiple databases so you can separate the public sites db from the tools database if you are extremely worried about security.