| 1. |
- Bohlin, Markus, et al.
(författare)
-
Bounding Shared-Stack Usage in Systems with Offsets and Precedences
- 2008
-
Ingår i: ECRTS 2008: PROCEEDINGS OF THE 20TH EUROMICRO CONFERENCE ON REAL-TIME SYSTEMS. - 9780769532981 ; , s. 276-285
-
Konferensbidrag (refereegranskat)abstract
- The paper presents two novel methods to bound the stack memory used in preemptive, shared stack, real-time systems. The first method is based on branch-and-bound search for possible preemption patterns, and the second one approximates the first in polynomial time. The work extends previous methods by considering a more general task-model, in which all tasks can share the same stack. In addition, the new methods account for precedence and offset relations. Thus, the methods give tight bounds for a large set of realistic systems. The methods have been implemented and a comprehensive evaluation, comparing our new methods against each other and against existing methods, is presented. The evaluation shows that our exact method can significantly reduce the amount of stack memory needed.
|
|
| 2. |
|
|
| 3. |
- Bohlin, Markus, et al.
(författare)
-
Safe Shared Stack Bounds in Systems with Offsets and Precedences
- 2008
-
Rapport (övrigt vetenskapligt/konstnärligt)abstract
- The paper presents two novel methods to bound the stack memory used in preemptive, shared stack, real-time systems. The first method is based on branch-and-bound search for possible preemption patterns, and the second one approximates the first in polynomial time. The work extends previous methods by considering a more general taskmodel, in which all tasks can share the same stack. In addition, the new methods account for precedence and offset relations. Thus, the methods give tight bounds for a large set of realistic systems. The methods have been implemented and a comprehensive evaluation, comparing our new methods against each other and against existing methods, is presented. The evaluation shows that our exact method can significantly reduce the amount of stack memory needed. In our simulations, a decrease in the order of 40% was typical, with a runtime in the order of seconds. Our polynomial approximation consequently yields about 20% higher bound than the exact method.
|
|
| 4. |
- Bohlin, Markus, et al.
(författare)
-
Shared Stack Analysis in Transaction-Based Systems
- 2007
-
Ingår i: Work in Progress Proceedings RTSS'07, Tucson, Arizona, USA. ; , s. 37-40
-
Konferensbidrag (refereegranskat)abstract
- In this paper, we present our ongoing work on shared stack analysis for hybrid (static and dynamic) scheduled fixed priority systems. We present two methods that extend our previous work to support stack analysis for the general tasks model with offsets where several transactions can share a common run-time stack. The aim of this work is to support stack analysis of a wider range of systems.
|
|
| 5. |
|
|
| 6. |
- Bucaioni, Alessio, et al.
(författare)
-
Demonstrator for modeling and development of component-based distributed real-time systems with Rubus-ICE
- 2013
-
Ingår i: Open Demo Session of Real-Time Systems.
-
Konferensbidrag (refereegranskat)abstract
- We present a demonstrator for modeling and development of component-based vehicular distributed real-time systems using the industrial model Rubus Component Model (RCM) and its development environment Rubus-ICE (Integrated Component development Environment). It demonstrates various stages during the development process of these systems such as modeling of software architecture, performing timing analysis, automatic synthesis of code from the software architecture, simulation, testing, and deployment.
|
|
| 7. |
- Bucaioni, Alessio, et al.
(författare)
-
From Modeling to Deployment of Component-Based Vehicular Distributed Real-Time Systems
- 2014
-
Ingår i: Proceedings, International Conference on Information Technology. - : IEEE. - 9781479931873 ; , s. 649-654
-
Konferensbidrag (refereegranskat)abstract
- We present complete model-and component based approach for the development of vehiculardistributed real-time systems. Within this context, we model and timing analyze these systems using one of the state-of-the-practice modeling and timing analysis techniques that is implemented in the existing industrial model the Rubus Component Model and accompanying tool suite. As a proof of concept, we conduct a case study by developing an intelligent parking assist system which is adistributed real-time application from the vehicular domain. The case study shows various stages during the development such as modeling of software architecture, performing timing analysis, simulation, testing, automatic synthesis of code from the software architecture, deployment, and execution.
|
|
| 8. |
- Carlson, Jan, et al.
(författare)
-
Deployment Modelling and Synthesis in a Component Model for Distributed Embedded Systems
- 2010
-
Ingår i: Proceedings - 36th EUROMICRO Conference on Software Engineering and Advanced Applications, SEAA 2010. - Lille, France. - 9780769541709 ; , s. 74-82
-
Konferensbidrag (refereegranskat)abstract
- We present an approach to combine model-driven and component-based software engineering of distributed embedded systems. Specifically, we describe how deployment modelling is performed in two steps, and present an incremental synthesis of runnable representations of model entities on various abstraction levels. Our approach allows for flexible reuse opportunities, in that entities at different levels of granularity and abstraction can be reused. It also permits detailed analysis, e.g., with respect to timing, of units smaller than a whole physical node. We present a concept, virtual nodes, which preserves real-time properties across reuse and integration in different contexts.
|
|
| 9. |
- Carlson, Jan, et al.
(författare)
-
Event-Pattern Triggered Real-Time Tasks
- 2008
-
Konferensbidrag (refereegranskat)abstract
- We present the concept of pattern-triggered tasks which are released when a particular pattern of events occur. A formal event algebra is used to define complex triggering conditions for these tasks, and the detection of triggering conditions is performed within the system by code generated automatically from these definitions. The implementation of the algebra has many desirable features for resource constrained real-time systems, including bounded and low execution time and memory consumption. Furthermore, we present novel schedulability analysis for our pattern-triggered tasks that leverage on existing analysis for fixed-priority and dynamic-priority scheduling policies.
|
|
| 10. |
- Hänninen, Kaj, et al.
(författare)
-
Determining maximum stack usage in preemptive shared stack systems
- 2007. - 1
-
Ingår i: Proceedings of the 9th Real-Time in Sweden (RTiS'07), Västerås, Sweden. ; , s. 118-126
-
Konferensbidrag (refereegranskat)abstract
- This paper presents a novel method to determine the maximum stack memory used in preemptive, shared stack, real-time systems. We provide a general and exact problem formulation applicable for any preemptive system model based on dynamic (run-time) properties. We also show how to safely approximate the exact stack usage by using static (compile time) information about the system model and the underlying run-time system on a relevant and commercially available system model: A hybrid, statically and dynamically, scheduled system. Comprehensive evaluations show that our technique significantly reduces the amount of stack memory needed compared to existing analysis techniques. For typical task sets a decrease in the order of 70% is typical.
|
|
