Results 1 
6 of
6
Modeling Software Systems with Rejuvenation, Restoration and Checkpointing through Fluid Stochastic Petri Nets
 in Proceedings of PNPM ’99
, 1999
"... In this paper, we present a Fluid Stochastic Petri Net (FSPN) based model which captures the behavior of aging software systems with checkpointing, rejuvenation and selfrestoration, three well known techniques of software fault tolerance. The proposed FSPN based modeling framework is novel in ma ..."
Abstract

Cited by 9 (3 self)
 Add to MetaCart
In this paper, we present a Fluid Stochastic Petri Net (FSPN) based model which captures the behavior of aging software systems with checkpointing, rejuvenation and selfrestoration, three well known techniques of software fault tolerance. The proposed FSPN based modeling framework is novel in many aspects. First, the FSPN formalism itself, as proposed in [24], is extended by adding flushout arcs. Second, the three techniques are simultaneously captured in a single model for the first time. Third, the formalism enables modeling dependencies of the three techniques on various system features such as failure, load and time in the same framework. Further, our base FSPN model can be viewed as a generalization of most previous models in the literature. To demonstrate, we present a set of FSPNs which are simple modifications of the base model. These represent software systems with checkpointing only, rejuvenation only and checkpointing and rejuvenation. We show that these FSPNs ca...
Jump Transitions in Second Order FSPNs
 In Proc. of MASCOTS’99
, 1999
"... In this paper jump transitions for the fluid model part in second order fluid stochastic Petri nets (FSPNs) are introduced. The extended formalism is defined, and the underlying partial differential equations are derived. With a jump a certain amount of fluid is added to a fluid place or taken out a ..."
Abstract

Cited by 3 (1 self)
 Add to MetaCart
In this paper jump transitions for the fluid model part in second order fluid stochastic Petri nets (FSPNs) are introduced. The extended formalism is defined, and the underlying partial differential equations are derived. With a jump a certain amount of fluid is added to a fluid place or taken out at once. This amount, the jump height, is sampled from a probability distribution. The dynamics of a model are described by second order partial differential equations that include integrals. As examples the virtual waiting time (or unfinished work) in a queueing system and a performability model of a multiprocessor system are modelled. 1. Introduction Stochastic Petri nets are a modelling formalism that is well suited for performance and reliability evaluation of complex computer and communication systems. To make them applicable for systems that have continuous components, like liquid in manufacturing systems or time, fluid places have been added, that are emptied and filled at a continu...
Second Order Fluid Stochastic Petri Nets: an Extension of GSPNs for Approximate and Continuous Modelling
 Proc. World Congr. on Systems Simulation
, 1997
"... A Petri net formalism is presented that allows for mixed discrete and continuous stochastic models. The continuous part of the models consists of fluid places that are filled and emptied at random (normally distributed) rate. Fluid places can be used for continuous approximation of heavily loaded di ..."
Abstract

Cited by 2 (0 self)
 Add to MetaCart
A Petri net formalism is presented that allows for mixed discrete and continuous stochastic models. The continuous part of the models consists of fluid places that are filled and emptied at random (normally distributed) rate. Fluid places can be used for continuous approximation of heavily loaded discrete places in order to avoid state space explosion as well as for the modelling of continuous system components. The dynamics of a second order FSPN are described by second order partial differential equations. They can be solved by using an implicit CrankNicolson scheme. As an example a processing system with modulated arrival stream and service break downs is used. QoS measures such as waiting time, loss rate and throughput are calculated for both a discrete and a fluid model. 1 Introduction Stochastic Petri nets (SPNs) [1] are well suited for the modelbased performance and dependability evaluation of computer and communication systems. As complexity of the systems increases the sta...
Hybrid Modelling with Second Order Fluid Stochastic Petri Nets
 In Proc. Workshop on Parallel and Distr. RealTime Systems
, 1997
"... Stochastic Petri nets have been extended for the modelling of hybrid systems by adding fluid places, that are filled and emptied continuously at random (normally distributed) rate. The flow rates as well as the discrete transition rates may depend on the content a fluid place holds. The dynamics of ..."
Abstract

Cited by 2 (1 self)
 Add to MetaCart
Stochastic Petri nets have been extended for the modelling of hybrid systems by adding fluid places, that are filled and emptied continuously at random (normally distributed) rate. The flow rates as well as the discrete transition rates may depend on the content a fluid place holds. The dynamics of the model are described by second order partial differential equations, that can be solved by using an implicit CrankNicolson scheme. Reward measures such as the fluid throughput can be computed from the continuous part of the solution. As an example we present a model of a chemical reactor. 1 Introduction Stochastic Petri nets (SPNs) [1] are well suited for the modelbased performance and dependability evaluation of complex systems. So far they are used for systems that take on discrete states only. Continuous Petri nets where first constructed by splitting tokens into a large number of particles [4], but they serve only poorly for performance evaluation since they do not include stochast...
On Markov Reward Modelling with Second Order FSPNs
 Proc. 15th annual UK Performance Engineering Workshop
, 1999
"... In this paper second order fluid stochastic Petri nets (FSPNs) will be used for modelling reward in the performability model. Two variations of the performability model are presented. In the first one two fluid places are used, one of which represents the profit (reward) obtained by operating the sy ..."
Abstract

Cited by 2 (2 self)
 Add to MetaCart
In this paper second order fluid stochastic Petri nets (FSPNs) will be used for modelling reward in the performability model. Two variations of the performability model are presented. In the first one two fluid places are used, one of which represents the profit (reward) obtained by operating the system and the other one the buffer, that is approximated continuously. In the second model only one fluid place is used, which represents the costs (negative reward) arising due to repair. The costs increase continuously while the system is in state of repair (which is a rate reward in the model). Additional costs incur each time the buffer fails (which is an impulse reward in the model). With a numerical solution algorithm the distribution of the reward and its mean are computed.
An Efficient Heuristics for Minimum Time Control of Continuous Petri nets ⋆
"... Abstract: This paper considers the problem of controlling timed continuous Petri nets under infinite server semantics. The proposed control strategy assigns piecewise constant flows to transitions in order to reach the target state. First, by using linear programming, a method driving the system fro ..."
Abstract
 Add to MetaCart
Abstract: This paper considers the problem of controlling timed continuous Petri nets under infinite server semantics. The proposed control strategy assigns piecewise constant flows to transitions in order to reach the target state. First, by using linear programming, a method driving the system from the initial to the target state through a linear trajectory is developed. Then, in order to improve the time of the trajectory, intermediate states are added by means of bilinear programming.