Executable System Modelling for Defence Options Analysis

For large, software-intensive networked systems, predicting system performance metrics early in the design lifecycle can be an extremely difficult task.

The problem

Whilst the system's model components are known at an early stage in terms of functionality and high-level behaviour, they are usually not fully defined. Such an approach is needed, however in order to help identify, manage and retire risk associated with system design much earlier than traditional methods employed in the Australia defence sector.

The solution

In the laboratory the system-of-study is decomposed into components using middleware technologies: this middleware is deployed onto hardware similar to that found in a target platform, and the middleware is deployed using the same technologies that would be expected in the target platform; such as DDS and CCM. This is known as "System Execution Modelling". The rough initial approximations applied to the model are refined throughout the modelling phase through test and feedback on the concepts and help to better inform designers on the impact of their design choices. The deployment allows performance tests of an integrated system to be conducted very early in the design stage, highlighting problems that are usually only found toward the end of projects with “traditional” design methodologies.

CDCIN has participated in collaborative research and development with the Defence Science and Technology Organisation (DSTO) on a System Execution Modelling capability, where high-level system model components execute on representative hardware and software. The University of Adelaide and DSTO have established laboratories and models for early-stage testing of the performance of distributed computing systems such as maritime combat systems and avionics systems.

Further reading

CDCIN paper at MilCIS 2014 Session 2.7a: "Executable System of Systems Modelling for Options Analysis".

Tagged in Case study, CDCIN