J. A. Stankovic, "Real-Time Computing Systems: The Next Generation," in J. A. Stankovic and K. Ramamritham (Eds.), Tutorial: Hard Real-Time Systems, IEEE Computer Society, 1988, pp.14--37. 147  Home/Search   Document Not in Database   Summary   Related Articles   Check

....di erent conceptual and computational frameworks; 2. Providing a potential designer with a set of formal tools (languages, algorithms) simplifying and systematizing the design process itself; 2 The term predictability is used in this paper in the sense adopted by the real time community[45]: A system is predictable if and only if all the timing constraints in the system are provably satis ed. Fig. 1. The generic layered software architecture. 3. Supporting the design process with a set of software tools enabling easy prototyping of complex real time systems. The bottom layer, ....

J.A. Stankovic. Real-time computing systems: The next generation. In IEEE Tutorial on Hard Real-Time Systems, pages 14-37. Computer Society Press, 1988.

....to distributed systems with preemptable tasks. 1 Introduction Hard real time system design requires specification methods that allow easy comprehension by the human designer as well as formal exactness. Real time systems will be large and complex, their activities cooperating intensely [7]. To achieve acceptable confidence into the system s functional correctness and operation, it is inevitable that the designer has clear and easily accessible understanding of the system activities, their co operation, and constraints. The verifiable correctness of a system is of equal concern, ....

J. A. Stankovic. Real-time computing systems: The next generation. In IEEE Tutorial: Hard Real-Time Systems, pages 14--37. IEEE Computer Society Press, 1988.

....in Computer Controlled Systems Most computer controlled systems are also real time systems. Real time computing systems are defined as those systems in which the correctness of the system depends not only on the logical result of computation, but also on the time at which the results are produced (Stankovic, 1988). For instance, assume that one of the inputs of the computer system concerns an alarm condition. The computer system must be able to handle such alarm condition (process that input and produce outputs accordingly) within a bounded time span. Thus, a computer system must not only react to ....

Stankovic, J. (1988). Real-Time Computing Systems: the Next Generation. In Tuturial: Hard Real-Time Systems, Stankovic, J. and K. Ramamritham (Editors), IEEE Computer Society Press, Los Alamitos, USA, pp. 14-38.

....it is usual to map system timing requirements onto process deadlines. The issue of meeting deadlines therefore becomes one of process scheduling. The development of appropriate scheduling algorithms has been isolated as one of the crucial challenges for the next generation of real time systems (Stankovic, 1988). One scheduling method that is used in hard real time systems is based upon rate monotonic theory (Liu, 1973) At runtime a preemptive scheduling mechanism is used: the highest priority runnable process is executed. Priorities assigned to processes are inversely proportional to the length of ....

Stankovic, J.A. (1988). Real-Time Computing Systems: The Next Generation. COINS Tech. Rep. 88-06, Dept. of Comp. and Inf. Sci., Univ. of Massachusetts.

....adapted to encompass priority based message dispatching in P NET networks. 1. Introduction Real time computing systems are defined as those systems in which the correctness of the system depends not only on the logical result of computation, but also on the time at which the results are produced [1]. Distributed real time systems are those systems that have timeliness requirements and include several inter operating computing components. In distributed real time systems, the co operation between the different computational devices is supported by some kind of inter processor communication. ....

Stankovic, J.: "Real-Time Computing Systems: the Next Generation", in Stankovic J., Ramamritham, K. (Eds.) "Tutorial: Hard Real-Time Systems" (IEEE, 1988), pp. 1438, 1988.

....of the first come first served (FCFS) PROFIBUS queue implementations. 1 Introduction Real time computing systems are defined as those systems in which the correctness of the system depends not only on the logical result of computation, but also on the time at which the results are produced [1]. There are various examples of real time computing systems, ranging from distributed computer control to robotics. In this paper we particularly address distributed computer controlled systems (DCCS) applications. A recent trend in DCCS is to interconnect distributed elements by means of a ....

Stankovic, J.: "Real-Time Computing Systems: the Next Generation", in STANKOVIC J., RAMAMRITHAM, K. (Eds.) "Tutorial: Hard Real-Time Systems" (IEEE, 1988), pp. 14-38, 1988.

....Task execution time is a prime example of such a parameter. A scheduling model that ignores variability in parameters, runs the risk of a catastrophic breakdown during execution [SA00b] A second feature, peculiar to real time systems is the existence of complex constraints between tasks [Sta88, LTCA89, Das85, JM86]. Traditional models such as the ones proposed in [Cof76, DL78] and [Pin95, Bru81] used fixed values for process time. Likewise, relationships between tasks are restricted to precedence constraints. Our goals in this paper are twofold: ffl Proposing a model that captures the intricacies involved ....

J. A. Stankovic. Real-time computing systems: The next generation. In J. A. Stankovic and K. Ramamritham, editors, Tutorial: Hard Real Time Systems, page 14:38. IEEE, 1988.

....queries in this model. Our model is very general and applicable to diverse areas ranging from real time process scheduling in operating systems and avionics to manufacturing and traffic control. 1 Introduction In real time scheduling, we are often confronted with process parameter variability [SA00b, SA00d, Sta88]. Process execution time is a prime example of a variable parameter. Secondly, the execution of real time processes is constrained through relationships that exist between their start times ( and execution times ) SA00a, Das85, BFR71, GH93] Traditional models do not accomodate either feature ....

J. A. Stankovic. Real-time computing systems: The next generation. In J. A. Stankovic and K. Ramamritham, editors, Tutorial: Hard Real Time Systems, page 14:38. IEEE, 1988.

....t2 has priority over t1 and both transitions share a common input place. Thus, after firing t2, transition t1 is no longer enabled. Moreover, these two types of intervals are important when dealing with real time systems since, in this kind of systems, different timing constraints are defined [5, 21]. There are three general categories of timing constraints: maximum timing constraints, minimum timing constraints and durational timing constraints. The maximum timing constraints are related to the maximum time that may elapse between two events. P1 t1 t2 t3 P2 P3 E= 3,3] E= 1,1] E= 2,2] ....

J.A. Stankovic. Real-time computing systems: the next generation. In J.A. Stankovic and K. Ramamritham, editors, Tutorial: Hard Real-Time Systems, pages 14 -- 37. Computer Society Press of the IEEE, 1988.

....in a given time interval and for any single task is also derived. The optimal algorithm is shown to be particularly useful when schedules are built from the integral flow values obtained from the corresponding maximum flow network. 1 Introduction Hard real time systems are those systems [St88] in which the correctness of the system depends not only on the logical result of computation, but also on the time at which the results are produced. Severe consequences will result if timing correctness properties of the system are violated. Applications of real time systems can be found in ....

J.A. Stankovic, Real-Time Computing Systems: the Next Generation, in Tutorial on Hard Real-Time Systems, edited by J.A. Stankovic and K. Ramamritham, IEEE Press,

....distinct engineering disciplines. While it shares many aspects with software engineering, additional specifications arise. The differences between standard software engineering and real time systems has often been ignored or misconstrued. These misconceptions have been expressed by Stankovic in [20]. The concepts from that article are presented below. There is no Science in Real Time Systems: Real time systems are much more than simply ensuring timing constraints of single tasks. There are many difficult, poorly understood interactions that occur between complex components in an extremely ....

Stankovic, J. A.,"Real-Time Computing Systems: The Next Generation" IEEE Computer, 1988

....an air traffic control region and will cease to exist when the aircraft leaves the region. Thus, a scheduling problem in a hard real time system is defined by the model of the system, the nature of tasks to be scheduled, and the objectives of the scheduling strategy. Not very much is known (cf. [12]) about scheduling algorithms for real time systems. Most of the existing results either pertain to simplistic situations or single processor systems. For a survey of the existing results, the reader is referred to [3] In, fact, most of the scheduling problems of interest to practical real time ....

J.A. Stankovic, Real-Time Computing Systems: the Next Generation, in Tutorial on Hard Real-Time Systems, edited by J.A. Stankovic and K. Ramamritham, IEEE Press, 1988.

....framework for handling absolute and relative time aspects. A consensus glossary of temporal database concepts is given in [28] Our ontology of time aspects for workflow management as given in section two complies to this glossary. Scheduling in general is known to be NP complete (see [17] and [36]) Comprehensive surveys on real time databases can be found in [2] and [33] Our scheduling approach will follow some of the ideas of [2] especially using soft real time. Special scheduling algorithms for real time databases are e.g. detailed in [1] where earliest deadline, least slack and ....

J. A. Stankovic. Real--time computing systems: The next generation. In J. A. Stankovic, editor, Hard Real--Time Systems. IEEE Computer Society Press, 1988.

....other hand, soft real time systems are characterized by graceful degradation. 44 Stankovic defined real time computing as that type of computing where the correctness of the system depends not only on the logical result of the computation but also on the time at which the results are produced [89]. Dodhiawala et al. suggested that a realtime system should have the following four aspects of real time performance [15] speed of task execution, responsiveness, timeliness, and graceful degradation. Besides these, memory management also plays an important role in a practical real time system ....

J. A. Stankovic, "Real-Time Computing Systems: The Next Generation," in Tutorial in Hard Real-Time Systems, ed. K. Ramamritham, pp. 14-37, IEEE, 1988.

....for the system designer, formal tools and techniques must be derived. The last underlies many of the other requirements and will be accompanied only with breakthroughs in the theory of modeling real time systems. Real time systems have not attracted academic attention, Stankovic points out in [Sta88] due to some common misconceptions which are quoted here. 1. There is no science in real time design. 2. Advances in supercomputer hardware will take care of real time requirements. 3. Real time computing is equivalent to fast computing. 4. Real time programming is assembly coding, priority ....

John A. Stankovic. Real-time computing systems: The next generation. Tutorial: Hard Real-Time Systems, pages 14--37, 1988.

.... computational power without redesigning any of the nodes) survivability (to have several minimal paths increasing the survivability of the network in case of node link failures) and experimental flexibility (to emulate various architectures by disabling some of the links in the chosen topology) [Sta88b]. Fast, reliable and time constrained communications: most of the existing communications models may provide fast communications, but they do not know how long it will take to deliver a particular message. Hence, it is difficult to guarantee all real time messages. Architectural support for ....

STANKOVIC, J. "Real-Time computing systems: the next generation". En "Hard Real-Time Systems". Tech. Report TR-88-06, COINS Dept. University of Massachusetts. January 1988.

....a set of tasks to meet their prescribed deadlines, precedence constraints must be established, satisfied, and resources must be available in time for each task. Abrupt delays at any stage of the process can disrupt the system s behavior and objectives; i.e. delayed production of results [PAN93, STA 88a] Scheduling decisions are guided by various metrics that depend on the application domain. The variety of metrics suggested for real time systems indicates the different types of real time systems that exist in the real world, as well as the type of requirements imposed on them. Different ....

....together form the correctness criterion of any RTDB system. Therefore, management of time critical information through a database system requires the integration of concepts from both fields in order to properly handle timing constraints and data consistency [ABB 88b, BUC89, HUA89, RAM93, SIN88, STA 88a, ULU92, and ULU 95b] 2. Scheduling algorithms in the two systems, DBMS and real time operating systems; differ in their schedulable units. While tasks are considered as the schedulable unit in real time systems, transactions are the schedulable unit in DBMS. Tasks typically include an ....

[Article contains additional citation context not shown here]

J. Stankovic, "Real Time Computing Systems: The Next Generation", Tutorial Hard Real-Time Systems, ed. J. A. Stankovic, pp. 14-38, IEEE (1988).

....of a real time scheduler is to ensure tasks commence and terminate their executions to meet their specified deadlines. The fact that a task can be deprived of processor time at certain instances is usually of little consequence to a real time system, so long as its deadlines are not compromised [51]. To illustrate the difference in real time and non real time scheduling paradigms, assume a system is composed of the set of tasks shown in Table 3.1. Furthermore, assume that the deadline D of each task (the time by which the computation needs to be finished by) is equal to the task s period T. ....

J.A. Stankovic. Real-time computing systems: The next generation. In John A. Stankovic and Krithi Ramamritham, editors, Tutorial: Hard Real-Time Systems, chapter 2, pages 14--37. IEEE Computer Society Press, 1988.

....to the actuators. Most of the computer controlled systems are also real time systems. Real time computer systems are defined as those systems in which the correctness of the system depends not only on the logical result of computation, but also on the time at which the results are produced (Stankovic, 1988). For instance, assume that one of the inputs of the computer system is a Boolean information of an alarm condition. The computer system must be able to handle such alarm condition (process that input and produce outputs accordingly) within a bounded time span. Thus, a computer system not only ....

Stankovic, J. 1988. Real-Time Computing Systems: the Next Generation. Stankovic J., Ramamritham, K. (Eds.), IEEE Press. Tutorial: Hard Real-Time Systems: pp. 14-38.

....must develop sufficient understanding of the relevant software and hardware engineering principles. 1] Unfortunately, DPMA model curriculum does not satisfy this need. Because current trend in building RTS involves paradigms from mathematical, scientific, and engineering disciplines [14], it is our feeling that incorporating RTS into our curriculum can further enhance our students analytical capability. This course is part of a series of system courses we offer, e.g. distributed and networking systems, functional systems in business, etc. Based on our course material, we ....

J. A. Stankovic. Real-time computing systems: The next generation. In J. A. Stankovic and K. Ramamritham, editors, Hard Real-Time Systems, chapter 2, pages 14--38. IEEE Computer Society Press, 1988.

.... tasks have been included within the periodic scheduling framework through the use of aperiodic servers in [5, 6] Other modifications to the original periodic scheduling framework include adding synchronization [7] and scheduling for buses [8] A good overview of real time systems is found in [9]. A real time task set is composed of n tasks. Each task i ; 1 i n, is described by a period, T i , a deadline, D i , and a worst case estimated execution time, C i . We assume that all deadlines are before or at task periods. Each instantiation of a task, which occurs once per period, is ....

J. Stankovic, "Real-time computing systems: The next generation," IEEE Tutorial on Hard Real Time Systems, 1988.

....correctness by statically binding task executions to fixed slots via timelines. This ad hoc approach tended to result in brittle systems which were not only expensive to develop, but also extremely difficult and costly to upgrade and maintain. Recent advances in real time scheduling theory by [1, 2, 3, 4, 5, 6, 7, 8, 9, 10] and others have developed algorithmic scheduling solutions for real time computing that guarantee individual task execution times in multi tasking, interrupt driven environments. Unlike earlier timeline scheduling approaches, the scheduling theory ensures the timing correctness of real time tasks ....

J. Stankovic, "Real-Time Computing Systems: The Next Generation ", IEEE Tutorial on Hard Real Time Systems, 1988.

....10 x out of 10 instances (or with 100 x probability) For example, the guarantee level of a hard and soft real time transaction should be 1 and 0, respectively. Sec. 21.2 Characteristics and Requirements 506 21.2. 3 Predictability Real time computing is not equivalent to fast computing [29]. Rather than being fast, more important properties of RTDBS should be timeliness, i.e. the ability to produce expected results early or at the right time, and predictability , i.e. the ability to function as deterministically as necessary to satisfy system specifications, including timing ....

J. Stankovic. Real-Time Computing Systems: The Next Generation. Technical Report TR-88-06, University of Massachusetts, Amherst, January 1988. Also available as Misconceptioins about Real-Time Computing, IEEE Computer, October 1988.

....Time to make Real Time Computing Real Real time computing systems are and will continue to play an increasingly vital role in our world. Current real time systems are expensive to build and their properties are usually verified with ad hoc techniques, or with expensive and extensive simulations [Stankovic:88a]. Different components of such systems are extremely difficult to integrate with each other which adds to the cost and complexity of these systems. Minor changes in any of these components result in another round of testing and fixing [Zave:82] This brute force approach is not likely to scale up ....

....correct operation. In the past few years, different aspects of real time systems have been studied, namely: specification, languages, models and semantics. However, the absence of a unified formal framework that addresses the afforementioned issues severely limited the usefulness of these studies [Stankovic:88a], Joseph:88] In this paper, we present a single framework for real time systems which makes it possible to see the relation between specification, implementation, correctness, and performance issues. The framework we suggest is based on the Input Output Timed Automaton (IOTA read yota ) ....

[Article contains additional citation context not shown here]

John A. Stankovic, "Real-Time computing systems: The next generation", COINS Technical Report 88-06 -- University of Massachusetts Amherst, January 88.

....second. ffl Stability under transient overload. When the system is overloaded by events and it is impossible to meet al..l the deadlines, we must still guarantee the deadlines of selected critical tasks. Real time scheduling is a vibrant field. Several important research efforts are summarized in [15]. Among them, generalized rate monotonic (GRMS) theory is a useful tool that allows system developers to meet the above requirements by managing system concurrency and timing constraints at the level of tasking and message passing. In essence, this theory ensures that as long as the system ....

J. Stankovic. Real-time computing systems: The next generation. IEEE Tutorial on Hard Real Time Systems, 1988.

....determinacy. In this chapter, we present the design philosophy of the Maruti system, as well as discuss the design and implementation of Maruti II. 4. 1 Introduction Many complex, mission critical systems depend not only on correct functional behavior, but also on correct temporal behavior [2, 13]. These systems are called real time systems. The most critical systems in this domain are those which must support applications with hard real time constraints, in which missing a deadline may cause a fatal error. Due to their criticality, jobs with hard real time constraints must always execute ....

....remarks are presented in Section 4.7. 4.1.1 Design Goals The design of a real time system must take into consideration the primary characteristics of the applications which are to be supported. The design of Maruti has been guided by the following application characteristics and requirements [2, 12, 13]. Real Time Requirements. The most important requirement for real time systems is the capability to support timely execution of applications. In contrast, with many existing systems, the next generation systems will require support for hard, soft, and non real time applications on the same ....

J. A. Stankovic. Real-Time Computing Systems: The Next Generation. In J. A. Stankovic and K. Ramamritham, editors, Tutorial: Hard Real-Time Systems, page 14:38, IEEE, New York, 1988.

....it is usual to map system timing requirements onto process deadlines. The issue of meeting deadlines therefore becomes one of process scheduling. The development of appropriate scheduling algorithms has been isolated as one of the crucial challenges for the next generation of real time systems (Stankovic, 1988). One scheduling method that is used in hard real time systems is based upon rate monotonic theory (Liu, 1973) At runtime a preemptive scheduling mechanism is used: the highest priority runnable process is executed. Priorities assigned to processes are inversely proportional to the length of ....

Stankovic, J.A. (1988). Real-Time Computing Systems: The Next Generation. COINS Tech. Rep. 88-06, Dept. of Comp. and Inf. Sci., Univ. of Massachusetts.

....8: References to find a detailed reference for a particular reading. For more information about homework assignments, the reader should refer to Section 4: Assignments and Student Performance Evaluation. Table 3 1: Summary of Lecture Topics and Readings Lecture Topics Readings 1 Introduction [59] 2 Real Time Project Formulation: Software Life Cycle Models Lecture Notes 3 Real Time Requirements and Specification Methods [11, 29: Chapter 5] 4 Real Time Requirements and Specification Methods Continued Add. Ref. 2, 3, 13, 17, 32, 43, 65, 67] 5 Scheduling Issues [42, 61, 63] 6 Case Study: ....

Stankovic, J. A. (1988). Real-Time Computing Systems: The Next Generation. Tutorial: Hard Real-Time Systems. Stankovic, J. A.; and Ramamritham, K., eds., IEEE Computer Society Press, pp. 13-38.

No context found.

J. A. Stankovic, "Real-Time Computing Systems: The Next Generation," in J. A. Stankovic and K. Ramamritham (Eds.), Tutorial: Hard Real-Time Systems, IEEE Computer Society, 1988, pp.14--37. 147

No context found.

Stankovic,J.A. (1988). Real-Time Computing Systems: The Next Generation. In: Stankovic, J.A. and Ramamritham, K. (Ed.), Tutorial: Hard Real-Time Systems, 14--37. IEEE Computer Society Press.

No context found.

Stankovic, J. A. (1988b). Real-time computing systems: The next generation. In Stankovic, J. A., editor, Hard Real-Time Systems. IEEE Computer Society Press.

No context found.

J.A. Stankovic, "Real-time Computing Systems: The next generation," Tech. Report, Dept. of Computer and Information Science, Univ. of Massachusetts, USA, 1988.

No context found.

Stankovic, J. A. (1988b). Real-time computing systems: The next generation. In Stankovic, J. A., editor, Hard Real-Time Systems. IEEE Computer Society Press.

No context found.

J. A. Stankovic, "Real-Time Computing Systems: The Next Generation," in Tutorial in Hard Real-Time Systems, ed. K. Ramamritham, pp. 14-37, IEEE, 1988.

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