In this paper we discuss a set of design tools aimed at the scheduling analysis of distributed real-time systems. The toolset consists of a designer, a schedulability analyzer and an emulator. The designer allows the description of both the software architecture (at the concurrency abstraction level) and the hardware architecture, in terms of the basic objects that are relevant to real-time analysis. The user can specify timing constraints of the most general form. The tool assists in the binding of the software components into the hardware architecture, the selection of the resource management policies assigned to each hardware and software component and the (possible) deadline partitioning process (where needed by the analysis). The schedulability analyser can eventually guarantee the schedulability of the application tasks and the emulator can simulate sample system runs in order to predict the average case behavior and understand the critical scheduling sequences. Finally, we discuss the possibility of providing UML extensions in order to perform the same kind of real-time analysis in UML-compliant modeling tools.
The DASE Design Environment for Distributed Real-Time Analysis
DOMENICI, ANDREA;
1999-01-01
Abstract
In this paper we discuss a set of design tools aimed at the scheduling analysis of distributed real-time systems. The toolset consists of a designer, a schedulability analyzer and an emulator. The designer allows the description of both the software architecture (at the concurrency abstraction level) and the hardware architecture, in terms of the basic objects that are relevant to real-time analysis. The user can specify timing constraints of the most general form. The tool assists in the binding of the software components into the hardware architecture, the selection of the resource management policies assigned to each hardware and software component and the (possible) deadline partitioning process (where needed by the analysis). The schedulability analyser can eventually guarantee the schedulability of the application tasks and the emulator can simulate sample system runs in order to predict the average case behavior and understand the critical scheduling sequences. Finally, we discuss the possibility of providing UML extensions in order to perform the same kind of real-time analysis in UML-compliant modeling tools.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.