Modeling and early validation of software architectures is a key concern for embedded applications. The Architecture Analysis and Design Language (AADL) is a textual and graphical language dedicated to design and analysis of architectural models of applicative software and its execution platform. AADL Inspector can process AADL v2.2 (SAE AS-5506C) as well as its Behavior Annex and Error Annex extensions. The rich semantics of AADL enable the specification of advanced assurance cases sharing a common representation of the system and involving a variety of analysis domains including static properties, real-time, safety and security. Thanks to its textual syntax, AADL is scalable and can be used as a first-class front end modelling language for designing large scale software intensive systems or as a pivot format to implement software development toolchains.
AADL Inspector is a light and standalone tool with an easy-to-use graphical user interface composed of three main parts:
- An AADL source file browser that allows for the definition of hierarchical projects and clear access to available libraries.
- A multi-files AADL text editor.
- A set of customizable tabs to activate and display the outcomes of processing tools.
The current version of AADL Inspector (v1.8) offers the following features:
- Import of AADL textual specifications (individual files or hierarchical projects)
- Syntactic analysis (aadlrev)
- AADL v2.2 (AS-5506C)
- AADL Behavior Annex v2.0 (AS-5506/3 Annex D)
- AADL Error Model Annex v2.0 (AS-5506/1A Annex E)
- AADL ARINC 653 Annex (AS-5506/1A Annex A)
- AADL FACE™ Annex (AS-5506/4 Annex F)
- AADL v1, v2.0 and v2.1 upwards compatibility
- Converts older AADL models into v2.2 (with a few restrictions on property associations)
- Import of FACE™, UML/MARTE, SysML or Capella models
- Import of AADL libraries from GitHub
- Real-Time and Deployment wizards
- Static rules checkers
- Turnkey integration of OCARINA syntactic and semantic analysers
- Customizable AADL rules checkers with LMP
- Declarative model metrics
- Automatic detection of the root system
- Automatic elaboration of the instance model
- Timing analysis
- Turnkey integration of the CHEDDAR v3.2 analysis kernel
- AADL to Cheddar model transformation
- Enhanced output format for post-processing
- VCD (Value Change Dump) file output of the scheduling static simulation
- Response time analysis
- Scheduling Aware end-to-end Flow Latency Analysis (SAFLA)
- Event-based simulation
- Turnkey integration of the MARZHIN v2 Multi-Agents simulation engine
- Emulation of the AADL run-time (multi-core, multi-processor and multi-partition architectures)
- Display of dynamic time-lines for Processors, Buses, Processes, Threads and Shared Data
- Asynchronous user interaction: in and out events and data
- Simulation scenarios (input ports) and probes (output ports)
- VCD (Value Change Dump) file output of the simulation trace
- Model Properties spreadsheet
- Extracts main RT properties from all the Thread instances
- Software to Hardware allocation
- Safety analysis
- Generation of OpenPSA description file from AADL Error Annex
- Bridger to launch Arbre Analyste for Fault Tree Analysis
- Security analysis
- Customizable implementation of security policy rules
- Definition and verification of security rules (LAMP)
- PDF documentation generator
- Ada and C code generators (Ocarina)
- Tool customization capabilities
- “plug and check” analysis tools
- Scripting language for plugins definition
- Command line options
- Powered by LAMP (Logic AADL Model Processing)
- AADL and XML/XMI parsing
- Model query language
- Model constraints language
- Model transformation language
- Embedded rules in AADL specifications (LAMP Annex)
For a more information on the AADL Inspector product please refer to the Ellidiss Technologies website (www.ellidiss.fr).
for further information see AADL Inspector.
What's new in AADL Inspector 1.8 ?
What’s new in AADL Inspector 1.8 ?
Release 1.8 of AADL Inspector includes an update of many existing internal features and external tools. The main changes are listed below:
- Update of Cheddar to release 3.2
- Update of the OpenPSA Fault Tree generator
- Major update of the LAMP plugin:
- New Scheduling Aware end-to-end Flow Analysis (SAFLA) plugin, written with LAMP.
- New Security rules checker, written with LAMP.
- New FACE™ to AADL import, written with LAMP.
- New customizable SysML to AADL import, written with LAMP.
- New LAMP laboratory to experiment complex assurance case tools and multi-model processing
Please look at the AADL Inspector 1.8 Quick Start Guide and User Manual for further details.
FACE is a trademark of The Open Group
What about the SysML to AADL import feature ?
What about the SysML to AADL import feature ?
Most of existing SysML to AADL transformation tools are based on the definition of a dedicated UML profile to represent AADL constructs. This approach seems natural; however, it forces the system engineer to “think” AADL and requires SysML tool specific customizations.
On the contrary, our solution takes plain SysML XMI input files and applies LAMP transformation rules to build an AADL model. Source code of the transformation template is provided and can be customized to fit corporate or project specific SysML to AADL mapping.
A similar approach can be applied to any other model transformation.
What is LAMP ?
What is LAMP ?
The LAMP Laboratory included in AADL Inspector 1.8 is a powerful tool that enables the specification of customized processing rules and the implementation of complex assurance cases. LAMP stands for Logical AADL Model Processing and consists in giving access to standard Prolog programs inside AADL annex subclauses associated with AADL packages and components. Ellidiss Technologies provides a set of tools to parse AADL and XML/XMI based languages as well as Prolog libraries containing model accessors and utilities to help the user to develop his own exploration, constraints, transformation, and architectural reasoning rules. Use of an existing standard formal and declarative language such as Prolog brings almost unlimited model processing possibilities.