Efficient simulation of network performance by importance sampling

Heegaard, Poul E.

doi:10.1016/s1388-3437(97)80065-0

Cited by 8 publications

(8 citation statements)

References 32 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The idea of adaptive biasing was introduced in [Heegaard 1997] for multiclass, homogeneous demand networks, where an arriving call requires the same number of units among the resource types in its demand set. In Heegaard [1998], this method was applied to nonhomogeneous demands, and calls with different priorities.…”

Section: Adaptive Issc Methodsmentioning

confidence: 99%

See 1 more Smart Citation

An adaptive approach to accelerated evaluation of highly available services

Hsu

Devetsikiotis

2007

ACM Trans. Model. Comput. Simul.

View full text Add to dashboard Cite

We motivate and describe improved fast simulation techniques for the accelerated performance evaluation of highly available services. In systems that provide such services, service unavailability events are rare due to a low component failure rate or high resource capacity. Using traditional Monte Carlo simulation to evaluate such services requires a large amount of runtime. Importance sampling (IS) has been applied to certain instances of such systems, focusing on single-class and/or homogeneous resource demands. In this article, we formulate highly available services as multiresource losstype systems, and we present two IS methods for fast simulation, extending to multiple classes and nonhomogeneous resource demands. First, for the cases in which component failure rates are small, we prove that static IS using the Standard Clock (S-ISSC) method exhibits the bounded relative error (BRE) property. Second, for estimating failure probabilities due to large capacity or fast service in systems that have nonrare component failure rates, we propose adaptive ISSC (A-ISSC), which estimates the relative probability of reaching each possible state of system failure in every step of the simulation. Using A-ISSC, IS methods which are proven to be efficient can be extended to multidimensional cases, while still retaining a very favorable performance, as supported by our validation experiments.

show abstract

Section: Adaptive Issc Methodsmentioning

confidence: 99%

“…In Heegaard [1996], methods were suggested for calculating f N (l ), which are based on a weight for each l that is a product of contribution and likelihood. In Heegaard [1997], a more complex algorithm was suggested. De Boer et al [2005], suggested using cross entropy as the guide to do the sampling.…”

Section: Adaptive Issc Methodsmentioning

confidence: 99%

An adaptive approach to accelerated evaluation of highly available services

Hsu

Devetsikiotis

2007

ACM Trans. Model. Comput. Simul.

View full text Add to dashboard Cite

show abstract

“…IS is well explained in Glynn and Iglehart [1989], Shahabuddin [1994], Heidelberger [1995], Sadowsky [1996], and the several other references given there. Application of IS to the simulation of communication systems is studied, for example, by Parekh and Walrand [1989], Chang et al [1994;, Bonneau [1996], Falkner et al [1999], and Heegaard [1998]. Chang et al [1994] derived an asymptotically optimal change of measure, based on the theories of effective bandwidth and large deviations, for estimating the probability p that a queue length exceeds a given level x before returning to empty, given that the queue is started from empty, for a single queue with multiple independent arrival sources that satisfy a large deviation principle.…”

Section: Introductionmentioning

confidence: 99%

Estimating small cell-loss ratios in ATM switches via importance sampling

L’Ecuyer¹,

Champoux

2001

ACM Trans. Model. Comput. Simul.

View full text Add to dashboard Cite

The cell-loss ratio at a given node in an ATM switch, defined as the steady-state fraction of packets of information that are lost at that node due to buffer overflow, is typically a very small quantity that is hard to estimate by simulation. Cell losses are rare events, and importance sampling is sometimes the appropriate tool in this situation. However, finding the right change of measure is generally difficult. In this article, importance sampling is applied to estimate the cell-loss ratio in an ATM switch modeled as a queuing network that is fed by several sources emitting cells according to a Markov-modulated ON/OFF process, and in which all the cells from the same source have the same destination. The change of measure is obtained via an adaptation of a heuristic proposed by Chang et al. [1994] for intree networks. The numerical experiments confirm important efficiency improvements even for large nonintree networks and a large number of sources. Experiments with different variants of the importance sampling methodology are also reported, and a number of practical issues are illustrated and discussed.

show abstract

“…In reliability models for instance, ε may represent a maximum failure rate of a component in the case of dynamic models (Shahabuddin 1994), or the reliability of a component in the case of static models. In queuing performance evaluation models, ε may be chosen as 1/B where B is the buffer size, so that the buffer overflow probability γ → 0 as ε → 0 (Juneja 1993, Heegaard 1998). …”

Section: Properties Of Rare Event Estimatorsmentioning

confidence: 99%

“…In particular, there has been a line of research of the asymptotic behavior of rare event simulation estimators when the rarity of the events goes to 0, which has led to define new concepts such as bounded relative error, asymptotic optimality (both concerning the size of the confidence interval relatively to the considered rare event), and more recently bounded normal approximation (looking at the coverage of the confidence interval). These concepts focus on the precision attained and the robustness of the simulation estimators (both are important features, see for example the discussion in (Heegaard 1998)), but do not take into account the computational times associated with them. We show in this paper that those properties are not defined generally enough to encompass some classes of estimators actually exhibiting a very nice behavior.…”

Section: Introductionmentioning

confidence: 99%

New Measures of Robustness in Rare Event Simulation

Cancela

Rubino

Tuffin

Proceedings of the Winter Simulation Conference, 2005.

View full text Add to dashboard Cite

Rare event simulation requires acceleration techniques in order to i) observe the rare event and ii) obtain a valid and small confidence interval for the expected value. A "good" estimator has to be robust when rarity increases. This paper aims at studying robustness measures, the standard ones in the literature being Bounded Relative Error and Bounded Normal Approximation. By considering the problem of estimating the reliability of a static model for which simulation time per run is the critical issue, we show that actually those measures do not validate the satisfying behavior of some techniques. We thus define Bounded Relative Efficiency and generalized bounded normal approximation properties of the two previous measures in order to encompass the simulation time. We also illustrate how a user can have a look at the coverage of the resulting confidence interval by using the so-called coverage function.

show abstract

Efficient simulation of network performance by importance sampling

Abstract: Simulation model Parallel and distributed simulation Hybrid techniques Rare event provoking Variance reduction Figure 1.2 : An overview of speed-up simulation techniques.

Cited by 8 publications

References 32 publications

An adaptive approach to accelerated evaluation of highly available services

An adaptive approach to accelerated evaluation of highly available services

Estimating small cell-loss ratios in ATM switches via importance sampling

New Measures of Robustness in Rare Event Simulation

Contact Info

Product

Resources

About