Abstract-In this paper, we present the BandWidth Inheritance (BWI) protocol, a new strategy for scheduling real-time tasks in dynamic systems, which extends the resource reservation framework to systems where tasks can interact through shared resources. The proposed protocol provides temporal isolation between independent groups of tasks and enables a schedulability analysis for guaranteeing the performance of hard real-time tasks. We show that BWI is the natural extension of the well-known Priority Inheritance Protocol to dynamic reservation systems. A formal analysis of the protocol is presented and a guarantee test for hard real-time tasks is proposed that takes into account the case in which hard real-time tasks interact with soft real-time tasks.Index Terms-Real-time scheduling, dynamic system, resource reservation, priority inheritance, constant bandwidth server. ae 1I NTRODUCTIONT HE main goal of a real-time scheduler is to provide temporal guarantees. In a hard real-time system, the scheduler must guarantee that, under certain worst-case assumptions, the temporal constraints of all tasks are respected. When mixing hard, soft, and non-real-time tasks, providing such temporal guarantees becomes a complex problem. In dynamic real-time systems, tasks can be activated dynamically and the system has no a priori knowledge about their run-time behavior. Classical real-time schedulability analyses are not appropriate for dynamic real-time systems because they require a priori knowledge of the characteristics of all the tasks to guarantee that every hard real-time task will meet its deadline.The resource reservation framework is a class of techniques that have been proven very effective in jointly scheduling hard real-time (HRT) and soft real-time (SRT) tasks. In particular, these approaches provide 1) temporal isolation between tasks and 2) schedulability analysis for HRT tasks. However, tasks are assumed to be independent. This is a severe limitation that hinders their utilization in real operating systems. The extension of the underlying model to cope with tasks that access shared resources through mutually exclusive (mutex) semaphores has only recently been addressedIn this paper, a new protocol, BandWidth Inheritance (BWI), is presented. It extends the Constant Bandwidth Server (CBS) algorithm [4] to real-time tasks that can access shared resources via critical sections, by using a technique derived from the Priority Inheritance Protocol (PIP) [5].In the remainder of this paper, we describe the BWI protocol and its properties. Then, we provide a schedulability analysis for HRT tasks. This protocol does not require any a priori knowledge of each task's behavior. Only the analysis is based on the knowledge of the worstcase behavior of the HRT tasks. Hence, the BWI protocol is suitable for a real operating system. 2R ELATED WORKDeng and Liu [6] proposed a model of dynamic real-time systems called the open system model. In their model, an application is a set of tasks and a server (i.e., an algorithm of the class...
This paper presents an object-oriented software tool, called RTSIM, aimed at simulating real-time embedded controllers. The tool consists of a collection of C++ libraries permitting a separate specification of the functional behaviour of the controller and of the hardware/software architecture to be used for its deployment. In particular, it is possible to provide an accurate modelling of the concurrent architecture of the control tasks and of the run-time support offered by the operating system for the real-time scheduling of the shared resources (CPU, memory buffers and network links). In this way, it is possible to compare different scheduling solutions by evaluating their simulated performance directly in the domain of the control application. Moreover, the tool can be utilized to tune up design parameters such as the activation frequencies of the tasks. The application of the tool is shown in a meaningful case study.The integration of flows of data from heterogenous sensors, having different requirements in terms of sampling rates and computation times, almost naturally induces concurrent implementation schemes. The ability of the system designer to specify, manage, and verify the functionality and performance of real-time concurrent processes (tasks) turns out to be a crucial success factor. Moreover, in the design of mass-produced embedded systems, the choice of hardware has a strong influence on the economy of the solution. Therefore, even in front of increasingly complex problems, the push towards minimization of computing hardware cost remains a dominant factor. In this context, an increasing emphasis is put on the effectiveness and on the efficiency of the production process of real-time software. Traditional development cycles tend to 'rigidly' separate the work of control engineers from that of software engineers, but the final outcome is often far from optimal in terms of performance/cost criteria.In order to introduce a profound innovation in this field, the availability of co-design tools spanning diverse engineering disciplines is of utmost importance.This paper focuses on one of the most familiar problems in real-time control software design, i.e. how the performance of a controller is affected by architectural and implementation choices (e.g. the decomposition of feedback controllers into tasks, the allocation of computation resources to tasks, the scheduling of the shared resources, etc.). Realistic and quantitative answers to this question during the early phases of the development are precious tools for product development.The concept of performance evaluation for a real-time controller can be developed along different directions. Most of the research in the area of real-time computing has studied the performance of concurrent software systems under the viewpoint of their timing behaviour. Ever since the seminal work of Liu and Layland [1], a fundamental performance metric is considered to be the tasks' schedulability, i.e. the ability for a set of tasks to execute respecting their assigned ...
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.