A. Bobbio, A. Puliafito, and M. Telek. A modeling framework to implement preemption policies in non-Markovian SPN. IEEE Transactions Software Engineering, 26:36--54, 2000.  Home/Search   Document Details and Download   Summary   Related Articles   Check

....is restricted to remove one job from one queue and add one job to another queue, 7 along with several other properties, then a product form solution is possible. An important aspect of the functionality of the action is what we call the execution policy. We use the term execution policy, as in [34], for a set of rules to unambiguously define the underlying stochastic process that describes the behavior of a model in the Mobius framework. We review below the execution policy of the Mobius framework. Finally, like state variables, the action provides a common interface by which other model ....

....The execution policy allows us to implement in the Mobius framework all the various preemption policies: the preemptive resume (prs) preemptive repeat di#erent (prd) 8 Start : R # WE : 0, 1] MTE : 0, 1] Fig. 5. Action state and preemptive repeat identical (pri) described in [34]; the concept of reactivation found in SANs [6] and other execution policies not generally considered by others (see [35] for examples) Actions must be defined so that they can utilize this general execution policy. We provide the action definition in the following sections. A.2 Action State ....

A. Bobbio, A. Puliafito, and M. Telek, "A modeling framework to implement preemption policies in non-Markovian SPNs," IEEE Transactions on Software Engineering, vol. 26, no. 1, pp. 36--54, Jan. 2000.

....to has to be fired at least after 1 time unit, because it is enabled since the initial state. Conversely, transition tl will always be fired, independently from the firing date of transition to. Notice that, when a transition is disabled , its timed informations are reseted, like a prd policy [7] or an enabling memory policy [1] 3. What is multiple enabledness The multiple enabledness of a Petri net is a notion that only appears with the semantics of real parallelism or with interleaving semantics of timed systems. This notion refers to the fact that a transition has enough tokens to ....

....of their ages. 5. Multiple enabledness and stochastic PN The multiple enabledness has been studied for stochastic PN in a more general work on execution policies [1] that includes single multiple infinite server policies, transition selection, memory policy, interrupt and resume policy [2, 7, 12]. In this paper, we assume a infinite server semantics, and we do not discuss the points related to the behaviour for single enabledness (like memory policy) but only they consequences on the multiple enabledness. Notice that the multiple enabledness is strongly related to the memory of the PN. ....

A. Bobbio, A. Puliafito, and M. Tekel. A modeling framework to implement preemption policies on nonMarkovJan SPNs. IEEE Transations on Software Engineering, 26(1):364, January 2000.

....favorable paths within a differentiated services environment. In the same context, preemption can be used to implement various prioritized access policies as well as restoration policies following fault events [1] Preemption policies have also been recently proposed in other contexts. In [5], the authors developed a framework to implement preemption policies in non Markovian Stochastic Petri Nets (SPNs) In a computing system context, preemption has been applied in cache related events. In [6] a technique to bound cache related preemption delay is proposed. Finally, in the wireless ....

A. Bobbio, A. Puliafito, and M. Tekel, "A Modeling Framework to Implement Preemption Policies in Non-Markovian SPNs," IEEE Transactions on Software Engineering, vol. 26, no. 1, pp. 36--54, January 2000.

....and what happens when they occur. Execution policies have been studied for some time. Examples of execution policies include various preemption policies such as preemptive resume (prs, also called race with enabling memory) preemptive repeat different (prd) and preemptive repeat identical (pri) [4], as well as reactivation [12] which we review in detail below. While each of these execution policies is suitable for representing behaviors within the formalism for which it was intended, it is not suitable for a general modeling framework such as Mobius. In particular, the multi formalism, ....

....policies used in various formalisms. We begin with a brief survey of the major existing execution policies we considered in developing the Mobius execution policy. For consistency, we use the term action for the basic state change mechanism of all formalisms. The most commonly used policy is prd [4, 1]. With prd, an action will complete in some random time after it becomes enabled. If, in the interval between enabling and completion, it becomes disabled or otherwise interrupted, then when the action becomes enabled again, it must choose a new completion time and essentially start over. With ....

[Article contains additional citation context not shown here]

A. Bobbio, A. Puliafito, and M. Telek. A modeling framework to implement preemption policies in non-Markovian SPNs. IEEE Trans. Softw. Eng., 26(1):36--54, Jan. 2000.

....is restricted to remove one job from one queue and add one job to another queue, along with several other properties, then a product form solution is possible. An important aspect of the functionality of the action is what we call the execution policy. We use the term execution policy, as in [4], for a set of rules to unambiguously define the underlying stochastic process that describes the behavior of a model in the Mobius framework. We review describe the Mobius execution policy below. Finally, like state variables, the Mobius action provides a common interface by which other model ....

....means preserve worker effort, discard minimum task effort, and preserve the delay and effort functions. This execution policy allows us to implement all the various preemption policies preemptive resume (prs) preemptive repeat different (prd) and preemptive repeat identical (pri) described in [4], the concept of reactivation found in SANs [25] as well as other execution policies not generally considered by others (see [18] for examples of these) Mobius actions must be defined so that they can utilize this general execution policy. We provide the action definition in the following ....

A. Bobbio, A. Puliafito, and M. Telek. A modeling framework to implement preemption policies in non-Markovian SPNs. IEEE Trans. Softw. Eng., 26(1):36--54, Jan. 2000.

No context found.

A. Bobbio, A. Puliafito, and M. Telek. A modeling framework to implement preemption policies in non-Markovian SPN. IEEE Transactions Software Engineering, 26:36--54, 2000.

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