Software Engineering Institute | Carnegie Mellon University
Software Engineering Institute | Carnegie Mellon University

Real-Time Performance

Real-time performance analysis

Designers must satisfy stringent real-time performance requirements that call for increased hardware diversity and complexity of embedded software systems.

Consider, for example, how a cruise controller on an automobile maintains vehicle speed over varying terrain when it is engaged by the driver. When the brake is applied, the system must relinquish speed control until told to resume. The system must also steadily increase or decrease vehicle speed when directed. Examples are easy to spot in other domains, too—such as automated teller machines and stability control in an aircraft.

The SEI approach provides the information that system designers need to make choices that reduce worst-case latency and ensure the meeting of critical deadlines. Also, it supports the analysis needed to determine the end-to-end schedulability of the system as conceived. Furthermore, the tools and techniques that the SEI uses allow a designer to create a single system architectural model to

  • prevent miscommunication among stakeholders
  • support analysis of architecture tradeoffs
  • cut risk and cost through early analysis, incremental development, and automated integration
  • reduce time-to-market
Real-Time Analysis Concern SEI Approach Answer
Reduce worst-case flow latency and execution time YES Decision support for adding processors, using a faster processor, rewriting code, or lowering controller’s signal processing rate
Assure schedulability YES Support for end-to-end timing and latency analysis
Fidelity of model to real system YES Capture of architecture and execution behavior in one model that represents the final system

The SEI Toolkit for Real-Time Analysis

The SEI uses the Architecture Analysis and Design Language (AADL) to document system architecture and provide a platform for multiple analyses. AADL, an international industry standard, supports multiple analyses from a single architectural model, enables modeling and analysis throughout the life cycle, and provides analysis of runtime behavior such as availability, performance, and security.

The SEI offers the Open Source AADL Tool Environment (OSATE) plug-in for worst-case, end-to-end timing, and flow latency analysis. This tool validates the latency of flow implementations and compares it to the latency value for the flow specification. If the flow implementation latency value exceeds the value for the flow specification, an error is reported. This analysis plug-in includes

  • a syntax-sensitive text editor, with integrated error reporting
  • a parser and semantic checker for textual AADL with conversion into AADL XML
  • an unparser for AADL XML to textual AADL conversion
  • support for multi-enterprise development through a version control system interface

The AADL also can be used with

Read our information sheet on this analysis.