The open-source Distributed Real-time Embedded Analysis Method (DREAM) project

Key challenges in the design, development and analysis of distributed real-time embedded (DRE) systems include safe composition of system components, formal analysis of real-time properties, the systematic measurement of coverage by simulations, and the automated generation of directed test vectors. Model-based technologies help address these issues by enabling design-time analysis and providing the means for the rapid evaluation of design alternatives with respect to end-to-end QoS properties, predictability and performance measures before committing to a specific platform. The Distributed Real-time Embedded Analysis Method (DREAM) is an open-source tool and method for the real-time verification and performance estimation of DRE systems. The project focuses on the practical application of formal analysis methods to real-time middleware to automate the verification, development, configuration, and integration of middleware-based DRE systems. DREAM was recently integrated in the Scenery project by Fujitsu Laboratories of America to analyze real-time properties in multi-processor system-on-chip (MPSoC) designs.

Project website: http://dre.sourceforge.net

Performance Estimation of Multi-processor System-on-Chip (MPSoC) Designs

As more and more functionality is provided by modern embedded systems, designers need to address a growing number of QoS constraints that have traditionally been relevant to hardware design, such as real-time properties, throughput, and energy consumption. Existing design methodologies focus on software structure and syntax, and cannot guarantee QoS properties. Component-based software development is an emerging paradigm to enable software reuse by composition. Components are high-level abstractions of software, that allow designers to focus on the essential characteristics of embedded software. Component-based software development provides the means to build reliable software from reliable building blocks. The analysis of component-based software, however, is still in its infancy. Software engineers are often unfamiliar with formal methods, and the required costs for analysis can often be justified only in the context of mission-critical systems. We are currently investigating the application of formal methods for the performance estimation of embedded HW/SW designs.

Alderis@UCI

The Alderis project focuses on the application of Domain Specific Modeling Languages (DSMLs) and meta-modeling to specify a common semantic domain for the analysis of distributed real-time embedded (DRE) systems. The Alderis language has both a visual and textual syntax with formally defined semantics. Alderis models can be verified directly by the open-source DREAM tool.

Project website: http://alderis.ics.uci.edu