MontiArc

Distributed interactive systems have become of more and more importance in the last decades. The application areas of embedded and otherwise distributed systems are vastly growing and so is the size of the individual systems. Typically interactive systems are composed of smaller systems or components. Some examples of such components in different domains are

• sensors, control units, and actuators in automobiles exchanging data values and control signals,
• real-time software controlling actions of actuators depending on sensors' data,
• interaction between objects via message passing in object oriented software systems or
• messages transmitted between web services in cloud computing applications.

These systems consist of many actors or components which are mostly independent of each other. In most such systems the objective however can only be carried out by cooperation of the independent sub systems. This cooperation is achieved by communicating via messages that are passed on well defined interfaces between the components. These messages can be

• event signals, e.g., messages on a bus,
• continuous values measured by sensors and discrete event signals,
• simple messages for signaling or method invocations or
• complex data structures passed between software services.

Some observable challenges in the design and implementation of these systems are for example finding a suitable (logical) decomposition of the system to model and analyze, creating compatible interfaces that allow communication of the right components, specifying the behavior/observed message exchange of the components, etc.

For the task of modeling distributed interactive systems by capturing elements of their logical or physical distribution we have developed the domain specific language MontiArc. MontiArc is a textual language and comes with an eclipse integrated editor. To do analysis on MontiArc models it is extended with a simulation framework that can execute behavior implemented in Java and attached to MontiArc models in a declarative way.

Beside the simulation of architectures the language frameworks serves several comfort features that ease the modeling of architectures:

• autoconnect: parameterizable automatically connection of compatible ports
• autoinstantiate: simultaneous instantiation and definition of subcomponents
• direct connection of subcomponents connecting their compatible ports
• implicit naming for ports and subcomponents with unique types