Electronica

HRT-HOOD† A Structured Design Method for Hard Real-time Systems‡ A. Burns and A.J. Wellings Real-time and Distributed Systems Research Group, Department of Computer Science University of York, Heslington, York, YO1 5DD, UK Email: burns@minster.york.ac.uk, andy@minster.york.ac.uk

Keywords: Structured Design Methods, Schedulability Analysis, Dependability

ABSTRACT

Most structured designmethods claim to address the needs of hard real-time systems. However, few contain abstractions which directly relate to common hard real-time activities, such as periodic or sporadic processes. Furthermore, the methods do not constrain the designer to produce systems which can be analysed for their timing properties. In this paper we present a structured design method called HRT-HOOD (HardReal-Time Hierarchical Object Oriented Design). HRT-HOOD is an extension of HOOD, and includes object types which enable common hard real-time abstractions to be represented. The method is presented in the context of a hard real-time system life cycle, which enables issues of timeliness and dependability to be addressed much earlier on in the development process. We argue that this will enable dependablereal-time systems to be engineered in a more cost effective manner than the current practise, which in effect treats these topics as performance issues. To illustrate our approach we present a simple case study of a Mine Drainage Control System, and show how it can be designed using the abstractions presented in the paper. 1. Introduction The most important stage in the development of anyreal-time system is the generation of a consistent design that satisfies an authoritative specification of requirements. Where real-time systems differ from the traditional data processing systems is that they are constrained by certain non-functional requirements (e.g. dependability and timing). Typically the standard structured design methods do not cater well for expressing these types ofconstraints[Kopetz1991]. The objective of this paper is to present a structured design method which is tailored towards the construction of real-time systems in general, and hard real-time systems in particular. We use the term hard real-time systems to mean those systems which have components which must produce
¡¡¡¡¡¡¡¡¡¡¡¡¡¡¡¡¡                                    † HOOD is a trademark of the HOOD User Group ‡The work has been supported, in part, by the European Space Agency (ESTEC Contract 9198/90/NL/SF) and by the UK Defence Research Agency (Contract Number 2191/023). This paper has been accepted for publication in the Real-time Systems Journal.

-2timely services; failure to produce a service within the required time interval may result in severe damage to the system or the environment, and maypotentially cause loss of life (for example in avionics systems). Rather than developing a new method from scratch, the HOOD method is used as a baseline. The new method, called HRT-HOOD (Hard Real-Time HOOD), was designed as part of an European Space Agency (ESA) supported project. HOOD was chosen as the base-line because ESA currently recommend the use of HOOD for their systems development.However, we believe the ideas presented in the paper can be used to extend other common design methods such as Mascot[Simpson1986]. Although hard real-time systems can be designed using structured methods such as HOOD and Mascot, these methods lack explicit support for common hard realtime abstractions. Consequently, their use is error prone and can lead to systems whose real-time properties cannot beanalysed. HRT-HOOD, in contrast, constrains the system decomposition so that the final design in amenable to timing analysis using such techniques as fixed priority or earliest deadline scheduling. A design method cannot be presented in isolation but must be considered within the context of the overall system life cycle of which it is a part. Unfortunately the traditional system life cycle also...