K. Gresser. An event model for deadline verification of hard real-time systems. In Proceedings 5th Euromicro Workshop on Real-Time Systems, pages 118--123, Oulu, Finland, 1993.  Home/Search   Document Not in Database   Summary   Related Articles   Check

....scheduling with message deadlines (MEDF) It will be shown that applying MEDF to a network of communicating processes will lead called critical instant 1.3. THIS WORK S BACKGROUND: RAPID PROTOTYPING 5 to an EDF processing sequence of these processes. Thus, feasibility analysis algorithms like [Gre93b, Gre93a] SSRB98] or [Jef92] may be applied to prove the required timeliness. Unfortunately, there exists no known MEDF implementation in state of the art real time operating systems. Therefore, necessary run time system support has to be provided. Schedulability analysis requires all ....

.... tasks with shared resources are summarized in [SSRB98] Feasibility conditions for sporadic task sets with common resources are presented for example in [Jef92] and [CLB99] Event driven systems that are triggered by stimuli with sporadic or even burst behaviour are investigated in [Gre93a, Gre93b] Gresser s EDF scheduling theorems and his Event Stream Model for description of task stimuli is the basis for this work. Astonishingly, although scheduling theory for both approaches is equally sophisticated, tool support is given mostly for rate monotonic analysis. TimeWiz (TimeSys) ....

[Article contains additional citation context not shown here]

Klaus Gresser. An event model for deadline verification of hard real--time systems. In Proc. Fifth Euromicro Workshop on Real Time Systems, pages 118--123, Finland, June 1993. IEEE. 5, 13, 28, 76, 77

....t and the maximum number of events b which may occur in a time interval T . They are also called inner (t) and outer (T ) periods of the event stream. This burst model is used by Tindell [21] and Audsley [1] to extend the analysis of static priority scheduling to such event streams. Gresser [8, 7] uses a somewhat similar model for analysis of dynamic priority scheduling. It introduces a vector of sequential time intervals rather than a single interval. This gives structure to an event sequence which can be exploited in analyzing Earliest Deadline First (EDF) scheduling. Sporadic events ....

K. Gresser. An event model for deadline verification of hard real-time systems. In Proceedings 5th Euromicro Workshop on Real-Time Systems, pages 118--123, Oulu, Finland, 1993.

....b) within a given time interval (the outer period T ) This outer period may also jitter (J) If known, a minimum time distance (the inter arrival time t) between two successive events within a burst can be specified. Sporadic events are captured by the minimum inter arrival time t, only. Gresser [4] provides a more general model. It introduces vectors of sequential time intervals. However, the model is not applicable to most of the analysis approaches. In a recent publication [14] we have presented an approach to interface between different event models. Table 1 shows simple ....

K. Gresser. An event model for deadline verification of hard real-time systems. In Proceedings 5th Euromicro Workshop on Real-Time Systems, pages 118--123, Oulu, Finland, 1993.

....assume a periodic activation of tasks. Recent extensions of the mentioned work allow periodic activation with jitter, e.g. 20] and arbitrary deadlines [12] Sprunt et al. 21] analyze the influence of sporadic task preemption. Tindell presents an approach for task bursts [24] Gresser [7] and Thiele et al. 23] use more general activation models. They introduce a vector of sequential time intervals rather than a few parameters (period, jitter, etc. only, in order to calculate performance quanta for each task. Gresser uses this model for analysis of dynamic priority (EDF) ....

K. Gresser. An event model for deadline verification of hard real-time systems. In Proceedings 5th Euromicro Workshop on Real-Time Systems, pages 118--123, Oulu, Finland, 1993.

....analysis method which fits to the WCET analysis approach in [3] The following section will provide an overview of the basic foundations the method presented Section 3 is build upon. Section 4 compares the approach with previously available work. 2 Foundations 2. 1 Model of Embedding Process In [4] a method for describing the environment a realtime system is embedded in is presented. The release of tasks is characterised as event streams. Each event corresponds to one release of task i and ES(T ) returns the maximum number of events in interval T . Figure 1 depicts a sample event stream ....

K. Gresser, "An event model for deadline verification of hard real--time systems," in Proc. Fifth Euromicro Workshop on Real Time Systems, (Finland), pp. 118-- 123, IEEE, June 1993.

....analysis method which fits to the WCET analysis approach in [3] The following section will provide an overview of the basic foundations the method presented Section 3 is build upon. Section 4 compares the approach with previously available work. 2 Foundations 2. 1 Model of Embedding Process In [4] a method for describing the environment a realtime system is embedded in is presented. The release of tasks is characterised as event streams. Each event corresponds to one release of task and 522 returns the maximum number of events in interval . Figure 1 depicts a sample event ....

K. Gresser, "An event model for deadline verification of hard real--time systems," in Proc. Fifth Euromicro Workshop on Real Time Systems, (Finland), pp. 118-- 123, IEEE, June 1993.

....5 Design Process Figure 4 depicts the tool chain of the automated rapid prototyping design process realized in REAR. The specifi cation in SDL is annotated with a specification of the timing requirements using deadlines and a temporal description of the embedding system with event streams ([7]) Currently, the SDL model is partitioned manually by allocating SDL processes on the target architecture s HW and SW processing units according to the task classification model ( 5] For the software part, C code is generated using SDT s code generator CAdvanced, automatically including ....

K. Gresser. An event model for deadline verification of hard real--time systems. In Proc. Fifth Euromicro Workshop on Real Time Systems, pages 118--123, Oulu, Finland, June 1993. IEEE.

....An idea for codesign with periodic tasks or processes is found in [2] The algorithm extends data dominated codesign approaches by a reactive process model to support the design of heterogeneous systems. A real time verification based on schedulability analysis for aperiodic tasks is described in [7]. This approach calculates, whether all tasks in the model guarantee their deadlines. In contrast to [1] and [13] the algorithm only finds deadline violations and does not calculate any metric values for the next step of the optimizer. Another disadvantage to use this algorithm for HW SW codesign, ....

K. Gresser. An Event Model for Deadline Verification of Hard Real-Time Systems. IEEE Proceedings of the 5th Euromicro Workshop on Real Time Systems, Oulu, Finland, 1993

....An idea for codesign with periodic tasks or processes is found in [2] The algorithm extends data dominated codesign approaches by a reactive process model to support the design of heterogeneous systems. A real time verification based on schedulability analysis of sporadic tasks is described in [7]. This approach calculates, whether all tasks in the model guarantee their deadlines. In contrast markbegin A P1 t1 t3 P2 t2 t4 P3 markend B span(A,B) 100 ms SETUPreq DATAreq Figure 1: Example of a PMSC specification to [1] and [13] the algorithm only finds deadline violations and ....

K. Gresser. An Event Model for Deadline Verification of Hard Real-Time Systems. IEEE Proceedings of the 5th Euromicro Workshop on Real Time Systems, Oulu, Finland, 1993

.... this purpose, we propose a few new keywords annotated to the SDL system (summarized in Table 1) Event Streams (ES) characterize the timely behaviour of the embedding process, i.e. they are a formal means to express the maximum possible number of events i of a certain type within an interval a i [8]. Considering the occurrence of interrupts on the left side of Fig. 2, the resulting ES 2 is shown on the right. ES annotations have to be added to all stimulating external events i.e. SDL signals coming from the environment and to all stimulating internal events, i.e. to all timers. ....

....by evaluting the specified or measured execution times depending of the design phase to find the worst case path in the TPG. Figure 5 shows the RTAM, derived from the TPG of the MCS example (Fig. 4) Schedulability Analysis For the EDF based schedulability analysis presented by Gresser in [8], the Event Streams (ES) introduced in Section 2.2.2 have to be transformed into an Event Function E(I) expressing the number of events per interval I . Figure 6 shows E(I) derived from the sample ES in Figure 2. The C(I) Function is defined as maximum computation time requested and due within ....

K. Gresser. An event model for deadline verification of hard real--time systems. In Proc. Fifth Euromicro Workshop on Real Time Systems, pages 118--123, Oulu, Finland, June 1993. IEEE.

....are the increasing flexibility and the ability to realize more complex applications in an elegant way. A disadvantage of event driven system components is the fact, that the occurrence of a single event or the combination of events is not predictable in advance without an exact process model (Gresser, 1993; Thielen, 1994) If such a model does not exist, but events should still be used as a powerful method, the described real time algorithms, which adapt themselves to the actual load situation, are a good alternative to guarantee timing constraints. 2.2 Classes and Properties A number of ....

Gresser, Klaus (1993). An event model for deadline verification of hard real--time systems. In: Proc. Fifth Euromicro Workshop on Real Time Systems. IEEE.

....(WCET) or amount of computation has to be determined. ffl Next, the tasks can be classified and allocated to the different types of processing units (RTU, HPU) ffl The required number and performance of the processing units can be quantified by the schedulability analysis developed by Gresser [4, 3]. In contrast to analysis methods for time driven (periodic) systems where task deadlines are guaranteed by the construction of the time driven schedule [6] this method takes into account the timely behaviour of the technical process which stimulates the event driven system s tasks. Event streams ....

K. Gresser. An event model for deadline verification of hard real--time systems. In Proc. Fifth Euromicro Workshop on Real Time Systems, pages 118--123, Oulu, Finland, June 1993. IEEE.

....opposed to soft real time , hard real time means that a deadline miss may result in loss of lives and money. We do not want to test the timely response by trial and error , but rather verify the necessary properties of the reactive system by the schedulability analysis developed by Gresser [1, 2, 3]. In contrast to analysis methods for time driven (periodic) systems where task deadlines are guaranteed by the construction of the time driven schedule, this method takes into account the timely behaviour of the technical process which stimulates the event driven system s tasks. Event streams ....

Klaus Gresser. An event model for deadline verification of hard real--time systems. In Proc. Fifth Euromicro Workshop on Real Time Systems, pages 118--123, Oulu, Finland, June 1993. IEEE.

....design process requires both, the specification of the systems behaviour as well as a specification of the timing properties and constraints of the system s embedding environment. The latter is done with the help of event streams, their associated maximum response time and event interdependences [5], while we use SDL for the functional specification. The specification is followed by the allocation of the SDL processes to the available processing units, according to the task classification model and based on the maximum response times and an estimated computational complexity. After that, ....

K. Gresser. An event model for deadline verification of hard real--time systems. In Proc. Fifth Euromicro Workshop on Real Time Systems, pages 118--123, Oulu, Finland, June 1993. IEEE.

.... analysis for event driven hard real time systems [10] First, the characteristics of the embedding process have to be described, i.e. the timely behaviour of stimulating events using Event Streams (ES) ES describe the maximum possible number of events i of a certain type within an interval a i [11]. Considering the occurrence of interrupts on the left side of Fig. 2, the resulting ES 1 is shown on the right. This leads to an Event Function E(I) expressing the number of events per interval I (Fig. 3) Single tasks are characterized by their WCET c max;j and a deadline d j for the ....

Klaus Gresser. An event model for deadline verification of hard real--time systems. In Proc. Fifth Euromicro Workshop on Real Time Systems, pages 118--123, Oulu, Finland, June 1993. IEEE.

No context found.

K. Gresser. An event model for deadline verification of hard real-time systems. In Proceedings 5th Euromicro Workshop on Real-Time Systems, pages 118--123, Oulu, Finland, 1993.

No context found.

K. Gresser. An event model for deadline verification of hard real-time systems. In Proceedings 5th Euromicro Workshop on Real-Time Systems, pages 118--123, Oulu, Finland, 1993.

No context found.

K. Gresser. An event model for deadline verification of hard real-time systems. In Proceedings 5th Euromicro Workshop on Real-Time Systems, pages 118--123, Oulu, Finland, 1993.

No context found.

K. Gresser. An event model for deadline verification of hard real-time systems. In Proceedings 5th Euromicro Workshop on Real-Time Systems, pages 118--123, Oulu, Finland, 1993.

Online articles have much greater impact   More about CiteSeer.IST   Add search form to your site   Submit documents   Feedback  

CiteSeer.IST - Copyright Penn State and NEC