Showing papers in "Performance Evaluation in 1989"

TL;DR: A distributed program is modelled as a stochastic network of tasks related by their rendezvous requests, which is assumed to have maximum concurrency as if each task were executed on its own processor.

TL;DR: A variant of an M/G/1 queuing model with finite waiting room with infinite waiting room is studied and the waiting time distribution, blocking probability and general queue length distribution are derived.

TL;DR: The analytic method for entropy maximisation, subject to marginal mean value constraints, is applied to characterise new product-form approximations for arbitrary FCFS queueing networks with multiple server queues and general interarrival and service times under repetitive-service blocking.

TL;DR: This paper presents some fundamental equivalence properties that hold for F/J networks with blocking and proves a theorem that provides a test for equivalence and illustrates it with several examples.

TL;DR: This paper will expose queueing theorists and applied queueing analysts to the practical queueing analysis problems that arise in computer performance modeling.

TL;DR: The broad conclusion is that the interior stations (especially the center stations) should be given preference over the end stations for receiving an extra server in systems of finite queues in series when the queue capacities are small or even zero.

TL;DR: These results can be used in approximations, in the buffer allocation problem as well as to explain the behavior of closed queueing networks with finite buffers.

TL;DR: A mean-value queueing model that considers bus latency, shared memory latency and bus interference as the primary sources of performance degradation in the system is developed and parametric results indicate how performance is affected by one of the parameters that characterizes data sharing in the workload.

TL;DR: It is concluded that the analysis method used and the results derived in this paper may be useful in deriving similar bounds for these methods as well and some specific implementations of these methods may be supported by the same bounds derived for the buffer occupancy method.

TL;DR: An approximate technique for analyzing cyclic queueing networks with exponential servers and finite buffers based on a decomposition approach is presented and a fairly good estimate of the throughput and the average queue lengths is provided.

TL;DR: A recursive procedure is developed to computed the steady-state probability distribution of the number of transactions in the processor, in which the generating function approach is also utilized to compute the probability of the empty-and-idle state.

TL;DR: Viewing networks with reversible routing as truncated processes, some of the results presented in the context of single class networks are shown to hold with multiple classes and BCMP type nodes.

TL;DR: Equivalencies between open and closed models of queueing networks with finite buffers are investigated and bounds on the throughput of open networks are established.

TL;DR: Simulation models are used to characterize the delay distribution of voice packets in a single hop as well as in a multi-hop network environment and the trade-off between the number of speakers that can be multiplexed using speech activity detection technique and the delay performance are quantified.

TL;DR: A methodology is presented for deriving optimal database partitions and transaction assignment so as to minimize the average system response time and can take advantage of probabilistic routing to enhance its performance.

TL;DR: A queueing model is presented for the performance evaluation of the gating scheme, and various performance measures are derived using the theory of piecewise Markov process.

TL;DR: A computationally efficient version of an algorithm reported by Altiok and Perros for the analysis of feed-forward open exponential queueing networks with blocking with the same accuracy as the original algorithm is presented.

TL;DR: In this article, a simple upper bound on the system throughput and an error bound for its accuracy can be concluded for arbitrary closed exponential queueing networks such as those with finite capacities and blocking or dynamic routing.

TL;DR: This paper gives the exact solution to the more general case of polling systems in which each station has a single buffer and shows that for the symmetric case the method requires the solution of 2N−1−1 equations, which is about one half of the number of equations required for the known solution.

TL;DR: The feasibility of the Shortest Time First (STF) approach to scheduling, in terms of predicted values obtained from other characteristics of the jobs, is investigated, and factors affecting its efficiency are identified.

TL;DR: Upper and lower stochastic bounds for a GI/G/1 vacation model with limited service are developed and the limited service vacation model is compared with the Bernoulli schedule vacation model.

TL;DR: A model that can analyze the average response time in round-robin systems is presented, based on the state transition model, which has been used to construct theaverage response time analyzer that is used as a postprocessor of the PASS system to determine a program's average responseTime.

TL;DR: An M/G/1 queueing system with two priority classes of traffic and Bernoulli feedback for the low priority class is analyzed and the distributed Poisson approximation is used to calculate the corresponding quantities when the high priority customers arrive at clock ticks.

TL;DR: It is shown that the control which achieves the maximum average voice packet coding rate under an average queue length (or an average waiting time) constraint has a simple structure specified by two parameters and is a randomized modification of two feedback controls with input saturation.

TL;DR: It is proved that the proposed model for circuit-switched networks has a product-form equilibrium state distribution which is insensitive to the forms of the call holding-time, failure-time and repair-time densities, thus extending results of Burman, Lehoczky and Lim.

TL;DR: It is conjecture that a service centre with moving server overhead could be used to represent one of these LAN algorithms in a product form queueing network model of a distributed system without introducing significant approximation errors.

TL;DR: The two approaches to aggregation considered represent two extensions of the idea of a Norton equivalent to non-product-form networks, one based on an exact result, the other heuristic, which are of interest beyond the particular application considered here.

TL;DR: A speed up technique for the approximate iterative solution of open tandem networks of queues with finite buffers that relies on a back and forth sweep of the network at each iteration so as to accelerate the return of the blocking information to the beginning of thenetwork.

TL;DR: These approximate solutions provide chain throughputs, mean end-to-end delays, and server utilizations that are sufficiently accurate for the analysis and design of communication networks and possibly other distributed systems with a large number of customer classes.