Architecture
The above figure provides an overview of the Graphical Modeler (GM) and EVA Netmodeler Architecture. Graphical Modeler runs in the web browser as a Smalltalk Applet, supported by a small Squeak virtual machine. The virtual
machine will be downloaded from the Internet (or other site designated by your EVA Netmodeler administrator) the first
time you launch Graphical Modeler. On subsequent occasions, it will be loaded from disk cache on your workstation. It is relatively small and only takes a few moments to download and launch.
GM communicates with an EVA Netmodeler server over the network (Intranet, Internet..) using standard web protocols, including TCP/IP and HTTP. It receives data as XML data streams and sends updates back to EVA Netmodeler using a
messaging API. The EVA Netmodeler Server manages users, security, meta models and all model content. It is a multi
-user environment which supports many concurrent users. It will also take care of locking with respect to multiple users wanting to update the same content simultaneously. EVA Netmodeler
lives behind a standard web server (e.g. Microsoft IISTM, ApacheTM) and uses a standard SQL'92 compliant database (e.g. SQL ServerTM, DB2TM, OracleTM...) For
storage of meta models and content models. It also manages related knowledge assets as electronic documents in server directories. Dependent upon security permissions, it is possible to view all EVA Netmodeler Content in the GM
Environment and all models created from GM in the EVA Netmodeler native browser based interfaces - essentially there
is one common meta model and repository which is shared by all tools subject to configuration and user permissions.
|
