L. Sha and J. B. Goodenough, "Real-Time Scheduling Theory and Ada", IEEE Computer (April 1990).  Home/Search   Document Not in Database   Summary   Related Articles   Check

....algorithm is designed for periodic task sets only, much work has been done to modify the algorithm to include aperiodic tasks, tasks with a secondary level of criticality, and task sets with unknown sizes. The techniques used to modify rate monotonic theory to include these cases are surveyed in [SHA90]. The importance abstraction [STRA92] is a framework for describing scheduling policies by focusing on how importance each task is to the system. Every system has a goal and the tasks within the system are processed with the intent of meeting the system goal. A task within the system is viewed as ....

Sha, L. and Goodenough, J. B., "Real-Time Scheduling Theory and Ada," IEEE Computer, Vol. 23, No. 4, pp. 53-62 (April 1990).

.... response times for aperiodic tasks because of its Chapter 2 Previous and Related Work 9 ability to provide immediate service for aperiodic tasks (due to it s high priority) Slack stealing is a relatively new approach which offers improvements in average response times over the sporadic server [ShG92 and SpB96]. The Slack stealing algorithm is an optimal service method for the sporadic server based on the idea of stealing all the possible processing time from the periodic tasks. The amount of slack time available during a given interval is equal to the amount of time that can be used to execute ....

Sha L., Goodenough J. (1992); "Real-Time Scheduling Theory and Ada"; Mission-Critical Operating Systems; IOS Press 1992; Washington D.C., 1992, pg. 294-319

....is used for estimation purposes. Rate monotonic scheduling (RMS) theory addresses the problem of ensuring that independent periodic tasks are scheduled without violation of the associated tim ing constraints. The real time scheduling basis for our work is found in the following two theorems [Liu73, Sha90]. Feasibility Theorem [Liu73] A set of n independent periodic tasks scheduled by the rate monotonic algorithm will always meet its deadlines, for all task phasings, if CI Cn(21 n ) n 1 = U(n) T1 nr where Ci and T are the execution time and period of task xi respectively. The ....

L. Sha, J.B. Goodenough: "Real-Time Scheduling Theory and Ada", IEEE Computer, Vol. 23, No. 4, pp.

....are independent if their correct execution does not imply need for synchronization. Note that tasks can be independent even when they interchange data, so long as the interchange does not cause a task to block. The real time scheduling basis for our work is found in the following two theorems [Liu73, Sha90]. Feasibility Theorem [Liu73] A set of n independent periodic tasks scheduled by the rate monotonic algorithm will always meet its deadlines, for all task phasings, if where C i and T i are the execution time and period of task t i respectively. Critical Zone Theorem [Liu73] For a set of ....

....an expensive model of context switching among tasks. Although, in this way the optimality of the rate monotonic scheduling can not be guaranteed anymore, the final achieved results are most often fully satisfactory. Both our experiments and previously published theoretical and experimental results [Sha90] indicate that preemption during rate monotonic scheduling is rarely required and used, even when allowed. More importantly, we provide a mechanism for rescheduling of all tasks until a need for preemption is not eliminated. 4 Experimental Results Table 3 gives descriptions of 16 tasks used to ....

[Article contains additional citation context not shown here]

L. Sha, J.B. Goodenough: "Real-Time Scheduling Theory and Ada", IEEE Computer, Vol. 23, No. 4, pp. 53-62, 1990.

....# 1 (6) provides a necessary and su#cient condition for the schedulability of any set of processes. However, this is no implicit scheduling technique. Instead an explicit static schedule or explicit dynamic schedule has to be constructed which governs the execution of strips of processes (see [8]) 2 Processes with age constraints A di#erent type of time constraint forces that subsequent executions of processes must not exceed time bounds given by the application context. One type of contraint emerges from the requirement that computational processes should operate on an internal image ....

L. Sha and J. B. Goodenough. Real-time scheduling theory and Ada. IEEE Transactions on Computers, pages 53--62, April 1990.

....reservation model is suitable for a wide range of applications, but applications that require different reservation semantics might need a model that takes advantage of other scheduling algorithms and analyses. For example, an application might require generalized rate monotonic analysis [39,64,107,109], aperiodic servers with different replenishment policies [111,112,120] earliest deadline first scheduling [27,49,67] sporadic task scheduling [51] or deadline monotonic scheduling [7] The reservation system also depends of priority inheritance protocols for fixed priority scheduling ....

L. Sha and J. B. Goodenough. Real-Time Scheduling Theory and Ada. IEEE Computer, 23(4):53--62, April 1990.

....critical section s actual priority Figure 1. A simple time diagram over the lifetime of three processes process is already an inherited priority or the blocking process may itself be blocked by some other process. However, the protocol informally described by Sha, Rajkumar and Lehoczky [10] has an error in that they propose as rule of disinheritance: When a process exits a critical section, it resumes the priority it had at the point of entry into the critical section. This rule misses all cases where other higher priority processes become blocked (directly or indirectly) by the ....

L. Sha and J. B. Goodenough. Real-time scheduling theory and Ada. IEEE Trans. Comput., pages 53--62, April 1990.

....designs, captured in automated form, can be subjected to various forms of dynamic and static analysis. For example, Sha and Goodenough show how multitasking designs can be analyzed, using rate monotonic analysis, to determine whether timing requirements can be met under worst case conditions. [SHA90] Smith and Williams describe a means of generating queuing models from multitasking designs to enable estimation of response time under typical system loads. SMIT93] Pidd presents guidelines for deriving data driven, generic simulators for 69 specific domains. PIDD92] Dillon shows how to ....

L. Sha and J.B. Goodenough, "Real-Time Scheduling Theory and Ada," Computer, April 1990, pp. 53-62.

.... with utilization less than U meets the admission criterion associated with scheduling policy Xg: 19) The concept of worst case achievable utilization was first introduced in the context of rate monotonic scheduling [13] and has been widely used in the design and analysis of real time systems [12, 21, 22, 24, 34]. The worst case achievable utilization is related to the fundamental requirements of predictability and stability in real time environments. Knowledge of the worst case achievable utilization provides a measure of predictability because if the system utilization is below the worst case achievable ....

L. Sha and J. B. Goodenough. Real-time scheduling theory and Ada. IEEE Computer, 23(4):53--62, Apr. 1990.

....represent a simplified model of most practical realtime applications. This basic model may not be of much practical relevance if it cannot be extended to accommodate other requirements. Recently this model has been adapted and extended in many aspects by researchers in solving practical problems [8, 69, 72]. FT RMMS Scheduling Problem: a set of n tasks S = is given with for i = 1, 2, n, where are the computation times of the versions of task t i . and are the release time, deadline, and period of task t i , respectively. The question is to schedule the task set S using the minimum number ....

L. Sha and J.B. Goodenough, Real-Time Scheduling Theory and Ada, Computer, April 1990, 53-65.

....a co scheduling strategy. ffl Scheduling Mismatch due to heterogeneity of resource scheduling policies: The main problem due to the lack of co scheduling among resources is the scheduling mismatch 2 NT CSS stands for network cooperative scheduling server. 4 problem. Priority inversion problem [9, 21, 25] exists because of the difference in scheduling policies between the controlled and the controlling resources. For instance, in the case of disk bandwidth guarantee, the controlled resource is the disk bandwidth. Its controlling resource is the processor. The processor scheduler can assign the ....

Sha, L. and Goodenough, J. B. Real-Time Scheduling Theory and Ada. Computer. May 1990.

....without violation of the associated timing constraints. Processes are independent if their correct execution does not imply need for synchronization. Note that processes can be independent even when they interchange data between them or have a set of imposed control and timing constraints [Sha and Goodenough 1990]. Stankovic et al. Stankovic et al. 1995] provided an excellent survey of a number of real time A Methodology and Algorithms for the Design of Hard Real Time Multi Tasking ASICs 9 of 30 Preliminaries scheduling algorithms, their computational complexity, assumed timing and hardware models. We ....

....starting points for our approach to synthesis of area efficient hard real time systems. The task level performance estimation techniques used in our methodology are based on the following two theorems related to a particular scheduling discipline rate monotonic scheduling [Liu and Layland 1973] [Sha and Goodenough 1990] which we use to develop task A Methodology and Algorithms for the Design of Hard Real Time Multi Tasking ASICs 10 of 30 Preliminaries level timing estimation techniques. Feasibility Theorem [Liu and Layland 1973] A set of n independent periodic processes scheduled by the rate monotonic ....

[Article contains additional citation context not shown here]

Sha, L. and Goodenough, J.B. 1990. "Real-Time Scheduling Theory and Ada", IEEE Computer, 23, 4, 53-62.

.... periodic model with starting time [12] Mok Chen later considered the variation of execution time and proposed a multiframe model [15] Baruah et al. further introduced generalized multiframe model [1] The scheduling of different models has also been extensively analyzed in other literatures [4, 6, 10, 11, 17, 18]. In this paper, wetake a look at the modeling itself and try to organize different models and their feasibility difficulties into a uniformed structure. Some of the work in above literature are recited or reproduced to makethischapter a selfcontained unit. There is yet another type of models in ....

Sha, L., Goodenough, J.B.: Real-Time Scheduling Theory and Ada. IEEE Computer. 23(4) (1990) 53--62

.... capabilities checking, and event vectoring specified in the MetaH specification[2] Process scheduling is based on rate monotonic scheduling theory, where the tool automatically assigns priorities, transforms periods as necessary, and generates code to manage process dispatching and continuation[8]. Inter process and inter processor communication code is statically scheduled and executed as part of the generated dispatcher. Eventvectoring, and the starting and stopping of processes at mode changes, are also controlled by generated code. All this must be done in a way that is mode dependent, ....

Lui Sha and John B. Goodenough, "RealTime Scheduling Theory and Ada," IEEE Computer, April 1990.

....on how far apart any two consecutive executions can be. Scheduling algorithms have been designed to schedule tasks with specific types of timing constraints. Two scheduling models are used in many existing real time systems: the cyclic executive (CE) model [BS89] and the fixed priority (FP) model [SG90]. Using the CE model, applications are divided into a sequence of non preemptable tasks. Tasks are activated in a fixed order as set in the predefined cyclic schedule. The cyclic schedule is repeated forever and usually does not change at run time unless special actions are needed. The FP model ....

L. Sha, J.B. Goodenough, "Real-Time Scheduling Theory and Ada", IEEE Computer, pp. 53-62, April 1990

....theory [LIU73] as a method for scheduling many periodic tasks on a single processor such that the scheduling algorithm used to do this was optimal. Dhall and Liu extended this work into the multiprocessor environment in [DHAL78] The following discussion is drawn largely from Sha and Goodenough [SHA90], who present an excellent overview of the theory and recent 37 extensions which include aperiodic and sporadic tasks, as well as non independent task relationships. Rate monotonic scheduling theory in essence ensures that as long as the processor utilization of all tasks lies below a certain ....

....is also optimal with respect to meeting deadlines among all algorithms where priority assignment may be made during the run of the system. In the deadline driven scheduling algorithm, the priorities are assigned according to which task s deadline is nearest rather than by period length. In [SHA90], Sha and Goodenough discuss the use of rate monotonic theory for realtime scheduling in the Ada tasking model. However, there are certain drawbacks to the unabridged rate monotonic scheduling policy, namely that (1) a task s period is not inherently related to how critical the task is to the ....

Sha, L. and Goodenough, J. B., "Real-Time Scheduling Theory and Ada," IEEE Computer, Vol. 23, No. 4, pp. 53-62 (April 1990). 147

....one. Hence, we seek to minimize A as much as possible in designing a heuristic algorithm. Despite the fact that the rate monotonic algorithm has become very populous among practitioners in solving many practical problems, and among researchers in generalizing the original results (see, e.g. [21]) previous results on the RMMS problem have been limited and the lack of progress in the area has been notable. The best bound was 2.00, which was derived about a decade ago, when the use of multiprocessor was not very common. Now the employment of multiprocessors for many real time applications ....

....used in a number of applications. For example, it has been specified for use with Space Station on board software as the means for scheduling multiple independent task execution; it will be built into the on board operating system [9] Many Ada compilers also support this scheduling discipline [21]. More recently, this scheduling discipline is used to schedule video, audio, and data streams in a multimedia system. The rest of the paper is organized as follows: In Section 2, we present the task model upon which the scheduling problem is defined. In Section 3, the new heuristic algorithm ....

[Article contains additional citation context not shown here]

L. Sha and J. B. Goodenough. Real-time Scheduling Theory and Ada. IEEE Computer, pages 53--66, April 1990.

....because: # it allows the deadline of a process to be less than its period # it allows sporadic processes to be more easily modelled # it allows for better control over jitter. Although Ada 83 has been widely criticised for it support for real time, the limitations are well understood[19, 6, 7] and extensive changes have been made in Ada 9X to make the language more responsive to the needs of the real time community[16, 1] The approach adopted by the project has been to use a restricted subset of Ada (e.g. no dynamic tasking, no dynamic memory allocation, no recursion) and to extend ....

L. Sha and J. B. Goodenough, "Real-Time Scheduling Theory and Ada", IEEE Computer (April 1990).

....for real time applications. Simplex utilizes RMA and GRMS techniques to ensure predictable execution, including upgrades of software components during system operation. A full description of RMS is beyond the scope of this paper. For a general introduction, see the article by Sha and Goodenough [Sha 90] A practitioner s guide is available in the book by Klein [Klein 93] RMS Theory was applied to the Coordinated Prototype to ensure that the pendulums continue to meet al..l real time constraints in their operation even in the presence of faults. Under RMS, higher frequency tasks are assigned ....

Sha, L. & Goodenough, J. B. "Real-Time Scheduling Theory and Ada." IEEE Computer, April 1990: 53-62.

No context found.

L. Sha and J. B. Goodenough, "Real-Time Scheduling Theory and Ada", IEEE Computer (April 1990).

No context found.

Lui Sha and John B. Goodenough. Real-Time Scheduling Theory and Ada. IEEE Computer, 23(4):53-62, 1990.

No context found.

Sha, L. and Goodenough, J. B. "Real-Time Scheduling Theory and Ada". IEEE Computer Vol. 23, No. 4 (April 1990).

No context found.

L. Sha, and J.B. Goodenough. "Real-Time Scheduling Theory and Ada". IEEE Computer, Vol. 23, No. 4, April 1990.

No context found.

L. Sha and J. B. Goodenogh, `Real-time scheduling theory and Ada', IEEE Computer, April 1990, pp. 53-62.

No context found.

Sha, L. and Goodenough, J. B. Real-Time Scheduling Theory and Ada. Computer. May 1990.

First 50 documents  Next 50

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