Brinkley Sprunt, Lui Sha, and John P. Lehoczky. Aperiodic task scheduling for hard real-time systems. Real-Time Systems Journal, 1(1):27--60, June 1989.  Home/Search   Document Not in Database   Summary   Related Articles   Check

.... Even though all child reserves are periodic, they can have di erent initial phases which can produce back to back execution (double hit [16] as in Deferrable Server (DS) which serves aperiodic requests [7] This back to back execution can be avoided using the Sporadic Server (SS) technique [15]. In this paper, we analyze both the Deferrable Server and Sporadic Server replenishment schemes. 3.4 Two step Schedulability Test The schedulability test for HDM needs to be done in two steps: 1. The schedulability of a parent reserve itself: This is done by a grandparent reserve. The ....

Brinkley Sprunt, Lui Sha, and John Lehoczky. Aperiodic task scheduling for hard-real-time systems. Real-Time Systems, 1:27-60, 1989.

....real world complexities such as synchronizing tasks, end to end deadlines that span multiple tasks, modal tasks that change characteristics at run time, and aperiodic tasks that may recur after an arbitrarily short time. These limitations and others have been resolved in the real time literature [17, 18, 21]. In contrast to the standard method for debugging systems software, a common approach for applying realtime analysis is: 1. select a system architecture, such as preemptive multithreading or a cyclic executive with interrupts; 2. select a task model, such as periodic tasks with precedence ....

Brinkley Sprunt, Lui Sha, and John P. Lehoczky. Aperiodic task scheduling for hard real-time systems. Real-Time Systems Journal, 1(1):27--60, June 1989.

....is improved with the proposed scheme. Another advantage of the proposed scheme is that the ceilings can be adjusted even when the complete information about the access patterns for the shared resources is not given. We assumed that most of soft deadline tasks are periodic. Many recent papers[7][8] 9] assumed that some soft deadline tasks are aperiodic, and tried to get faster response of soft deadline tasks. The ceiling adjustment in this paper can be easily extended to the algorithms that uses fixed priority periodic server for the aperiodic tasks, such as the polling server, ....

....tasks are aperiodic, and tried to get faster response of soft deadline tasks. The ceiling adjustment in this paper can be easily extended to the algorithms that uses fixed priority periodic server for the aperiodic tasks, such as the polling server, deferrable server, and the sporadic server in [7], and can get faster response times. The slack stealing algorithm is introduced in [8] and it is shown in [9] that PCP can be used with the slack stealing when soft and hard deadline tasks share the resources. To combine CAS with the slack stealing, the ceiling adjustment need to be modified ....

Brinkley Sprunt, Lui Sha and John Lehoczky, "Aperiodic Task Scheduling for Hard-Real-Time Systems," The Journal of Real-Time Systems, 1, pp. 27-60, 1989.

....overloaded conditions [12] while for dynamic algorithms it is almost impossible to predict which task will miss a deadline under such conditions. Extensions to the rate monotonic algorithm, such as the deferred server or sporadic server, makes it possible to handle aperiodic tasks as well [14]. Any future real time system will consist of periodic and aperiodic tasks and so it is unreasonable to leave this unconsidered. In a real time network system, aperiodic bursts of data are not uncommon. 5.2 RtP Server Mach s micro kernel architecture makes it possible for us to implement RtP as ....

....If an application does not receive any notification before this deadline, it can assume network failure or node crash, and take necessary actions. We are also considering the incorporation of the extended rate monotonic server algorithms such as the deferrable server or the sporadic server [14], to implement the CCH. 5.3 Application Process States Part of a state transition diagram of an application during the VDC establishment phase is shown in Figure 10. The white boxes depict messages sent from the application to the CCH, and the dark boxes show events that effect its state. By ....

Brinkley Sprunt, Lui Sha, and John P. Lehoczky. Aperiodic Task Scheduling for Hard-Real-Time Systems. Real-Time Systems, 1(1):27--60, June 1989.

....intelligent control to produce the best results it can under real time constraints and other resource (e.g. information, knowledge) constraints. Our conception of real time performance in intelligent agents is qualitatively different from conceptions embodied in other sorts of computer systems [3, 7, 16]. In particular, we do not view real time performance as a provable, guaranteed, universal property of the agent. Nor do we seek real time performance through effective engineering of the agent for narrowly specified task environments. We feel that these constructs are surely premature and ....

Brinkley, S., Sha, L., and Lehoczky, J. Aperiodic task scheduling for hardreal -time systems. Real-Time Systems, 1:1, 27-60, 1989.

....can act as if it has exclusive access to a uniformly slower processor. We will draw upon existing techniques for providing real time guarantees in the presence of high priority interrupt handlers [8] and for scheduling aperiodic tasks in systems where hard guarantees are given to periodic tasks [17]. 5 The Resource Manager New threads or threads that are changing mode make a request to the resource manager in order to obtain a guarantee for a particular type of scheduling. The resource manager is responsible for ensuring that the request is in conformance with userspecified policies about ....

Brinkley Sprunt, Lui Sha, and John P. Lehoczky. Aperiodic Task Scheduling for Hard Real-Time Systems. Real-Time Systems Journal, 1(1):27--60, June 1989.

....inscriptions for Fig. 27. 35 7.2.4 Aperiodic Servers Background Service Background service for aperiodic tasks can be modeled implemented simply by giving aperiodic tasks lower priorities than periodic tasks and using an unmodified fixed priority scheduler. Polling Server The polling server [46] for aperiodic tasks is released periodically like a periodic task. Every time it is released it services all pending aperiodic requests until its capacity is exhausted. If there are no pending aperiodic requests for that server at the time of its release, then it immediately suspends itself until ....

....removes and returns the capacity token. Also, for the allocation of capacity it is not only necessary 37 to check the available capacity of the server but also to check the remaining time of that server period. Priority Exchange Server With the priority exchange server scheduling algorithm [25, 45, 46, 11], the server is replenished at the beginning of each server period and pending aperiodics are served. Unused server capacity is conserved, but its priority is exchanged with the priority of the highest priority periodic task waiting for execution. This continues until the server priority has ....

Brinkley Sprunt, Lui Sha, and John Lehoczky. Aperiodic task scheduling for hard real-time systems. Real-Time Systems, (1):27--60, 1989.

....with tradeoffs in communication vs. performance and quality in distributed real time scheduling. In addition, research at Carnegie Mellon University has been extending priority based scheduling methods to address dynamic system behaviors, typically by development of novel scheduling algorithms[24]. Multiprocessor implementations of internally concurrent schedulers were first described in [7] then generalized and evaluated rigorously and experimentally in [23, 30] The latter work also developed new algorithms for schedulability analysis and addressed the scalability of schedulers in terms ....

Brinkley Sprunt, Lui Sha, and John Lehoczky. Aperiodic task scheduling for hard-real-time systems. The Journal of Real-Time Systems, 1:27--60, 1989.

....static scheduling algorithms, such as the rate monotonic algorithm and its variants, are geared towards scheduling periodic tasks. Tasks of a nonperiodic nature are not addressed. Various means of dealing with such tasks (within the framework of scheduling algorithms designed for periodic tasks) [50, 17, 47] are, in turn, described in this section. Before proceeding further, however, a distinction must be drawn between aperiodic tasks, which means that no indication is given as to when tasks may arrive in a system, versus sporadic tasks, which imply that a maximum of one instance of such a task may ....

....than it was initially allotted. This discrepancy in priorities makes the prediction of missed deadlines extremely difficult. Deadlines of sporadic tasks are most likely missed in a highly unpredictable fashion. CHAPTER 3. REAL TIME SCHEDULING 30 Deferrable Server The deferrable server method[31, 50], like the previous methods for dealing with sporadic or aperiodic tasks, allots a periodic server task that services sporadic or aperiodic events. Unlike the priority exchange server, should the deferrable server have no sporadic or aperiodic task to execute during its allotted computing time, it ....

[Article contains additional citation context not shown here]

Brinkley Sprint, Lui Sha, and John Lehoczky. Aperiodic task scheduling for hard real-time systems. Journal of Real-Time Systems, 1:27--60, 1989.

No context found.

Brinkley Sprunt, Lui Sha, and John P. Lehoczky. Aperiodic task scheduling for hard real-time systems. Real-Time Systems Journal, 1(1):27--60, June 1989.

No context found.

Brinkley Sprunt, Lui Sha, John Lehoczky. Aperiodic task scheduling for hard real-time systems. The Journal of Real-Time Systems, 1:27-60, 1989.

No context found.

Brinkley Sprunt, Lui Sha and John Lehoczky, "Aperiodic Task Scheduling for Hard-Real-Time Systems", Real-Time Systems, vol. 1, pp. 27-60, 1989.

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