Equilibrium Queue Research Articles

Response Time Distribution in a Tandem Pair of Queues with Batch Processing

Response time density is obtained in a tandem pair of Markovian queues with both batch arrivals and batch departures. The method uses conditional forward and reversed node sojourn times and derives the Laplace transform of the response time probability density function in the case that batch sizes are finite. The result is derived by a generating function method that takes into account that the path is not overtake-free in the sense that the tagged task being tracked is affected by later arrivals at the second queue. A novel aspect of the method is that a vector of generating functions is solved for, rather than a single scalar-valued function, which requires investigation of the singularities of a certain matrix. A recurrence formula is derived to obtain arbitrary moments of response time by differentiation of the Laplace transform at the origin, and these can be computed rapidly by iteration. Numerical results for the first four moments of response time are displayed for some sample networks that have product-form solutions for their equilibrium queue length probabilities, along with the densities themselves by numerical inversion of the Laplace transform. Corresponding approximations are also obtained for (non-product-form) pairs of “raw” batch-queues—with no special arrivals—and validated against regenerative simulation, which indicates good accuracy. The methods are appropriate for modeling bursty internet and cloud traffic and a possible role in energy-saving is considered.

Functional large deviations for Cox processes and $\mathit{Cox}/G/\infty$ queues, with a biological application

We consider an infinite-server queue into which customers arrive according to a Cox process and have independent service times with a general distribution. We prove a functional large deviations principle for the equilibrium queue length process. The model is motivated by a linear feed-forward gene regulatory network, in which the rate of protein synthesis is modulated by the number of RNA molecules present in a cell. The system can be modelled as a nonstandard tandem of infinite-server queues, in which the number of customers present in a queue modulates the arrival rate into the next queue in the tandem. We establish large deviation principles for this queueing system in the asymptotic regime in which the arrival process is sped up, while the service process is not scaled.

Recursive Solution of Queue Length Distribution for Geo/G/1 Queue with Delayed Min(N, D)-Policy

AbstractIn this paper we consider a discrete-time Geo/G/1 queue with delayed Min(N, D)-policy. Using renewal process theory, total probability decomposition technique and z-transform, we study the transient and equilibrium properties of the queue length from an arbitrary initial state, and obtain both the recursive expressions of the transient state queue length distribution and the steady state queue length distribution at arbitrary time epoch n+. Furthermore, we derive the important relations between equilibrium queue length distributions at different time epochs n–, n and n+. Finally, we give some numerical examples about capacity decision in queueing systems to demonstrate the application of the analytical results reported in this paper.

Managing service capacity with boundedly rational consumers under social interactions

Purpose Consumers are inclined to join longer queues due to social interactions in service consumptions. This purchase behavior brings in operational challenges in terms of capacity planning, which affects consumers’ demand, leading to an unstable and fluctuated arrival process. This paper aims to investigate the dynamic characteristics of the arrival process of a service system with boundedly rational consumers whose purchase decisions are influenced by the queue length under social interactions. Design/methodology/approach Consumers’ bounded rationality is modeled based on the random utility theory. Due to social interactions, the equilibrium queue length and its interaction with the expected waiting time affect consumers’ value perception. The authors first analyze the optimal service capacity decision with or without considering the influence of social interactions in a static setting. They then focus on the dynamic characteristics of the arrival process by a one-dimensional dynamical model in terms of the arrival rate. Findings This paper finds that the service system can behave chaotic in terms of arrival rate dynamics under social interactions. The results highlight the dynamical complexity of a simple service system due to consumers’ behavioral factors and the influence of social interactions, which may be the critical drivers leading to fluctuated and uneven demand. Originality/value The findings demonstrate that due to consumers’ limited cognitive ability and the influence of social interactions, the demand to a service system can be stable, periodic or even chaotic in terms of the arrival process. This study provides an alternative explanation to the observed demand fluctuations in various service processes under the influence of social interactions, which is important for service providers to effectively manage service capacity to achieve a stable service process and improve operational efficiency.

Some analysis results associated with the optimization problem for a discrete-time finite-buffer NT-policy queue

The prime objective of this paper is to give some analysis results concerning the discrete-time finite-buffer NT-policy queue, which can be utilized to determine the optimal threshold values. By recording the waiting time of the leading customer in server’s vacation period, the model is successfully described as a vector-valued Markov chain. Meanwhile, depending on the special block structure of the one-step transition probability matrix, the equilibrium queue length distribution is calculated through a more effective UL-type RG-factorization. Due to the number of customers served in the busy period does not have the structure of a Galton-Watson branching process, analysis of the regeneration cycle is regarded as a difficult problem in establishing the cost structure of the queueing system. However, employing the concept of i-busy period and some difference equation solving skills, the explicit expression for the expected length of the regeneration cycle is easily derived, and the stochastic decomposition structure of the busy period is also demonstrated. Finally, numerical results are offered to illustrate how the direct search method can be implemented to obtain the optimal management policy.

Geo (λ)/ Geo (μ) +G/2 queues with heterogeneous servers operating under FCFS queue discipline

This article discusses the steady analysis of a discrete time queue of Geo/Geo+G/2 type. All arriving customers are served either by server-1 according to a geometrically distributed service time S1=k slots for k=1,2, …∞, with mass function f1(k)==Pr(S1=k) = μ(1- μ) k-1 with mean rate 0 or mean service rate μ2=1/β. Sequel to some objections raised on the use of the classical 'First Come First Served (FCFS)' queue discipline when the two heterogeneous servers operate as parallel service providers, an alternative queue discipline in a serial configuration of servers are considered in this work; the objective is that if, in a single-channel queue in equilibrium, the service rate suddenly increases and exceeds the present service capacity, install a new channel to work serially with the first channel as suggested by Krishnamoorthy (1968). Using the embedded method subject to different service time distributions we present an exact analysis for finding the ‘Probability generating Function (PGF)’ of steady state number of customers in the system and most importantly, the actual waiting time expectation of customers in the system. This work shows that one can obtain all stationery probabilities and other vital measures for this queue under certain additional and simple but realistic assumptions.

Decay of tails at equilibrium for FIFO join the shortest queue networks

In join the shortest queue networks, incoming jobs are assigned to the shortest queue from among a randomly chosen subset of $D$ queues, in a system of $N$ queues; after completion of service at its queue, a job leaves the network. We also assume that jobs arrive into the system according to a rate-$\alpha N$ Poisson process, $\alpha<1$, with rate-1 service at each queue. When the service at queues is exponentially distributed, it was shown in Vvedenskaya et al. [Probl. Inf. Transm. 32 (1996) 15-29] that the tail of the equilibrium queue size decays doubly exponentially in the limit as $N\rightarrow\infty$. This is a substantial improvement over the case D=1, where the queue size decays exponentially. The reasoning in [Probl. Inf. Transm. 32 (1996) 15-29] does not easily generalize to jobs with nonexponential service time distributions. A modularized program for treating general service time distributions was introduced in Bramson et al. [In Proc. ACM SIGMETRICS (2010) 275-286]. The program relies on an ansatz that asserts, in equilibrium, any fixed number of queues become independent of one another as $N\rightarrow\infty$. This ansatz was demonstrated in several settings in Bramson et al. [Queueing Syst. 71 (2012) 247-292], including for networks where the service discipline is FIFO and the service time distribution has a decreasing hazard rate. In this article, we investigate the limiting behavior, as $N\rightarrow \infty$, of the equilibrium at a queue when the service discipline is FIFO and the service time distribution has a power law with a given exponent $-\beta$, for $\beta>1$. We show under the above ansatz that, as $N\rightarrow\infty$, the tail of the equilibrium queue size exhibits a wide range of behavior depending on the relationship between $\beta$ and $D$. In particular, if $\beta>D/(D-1)$, the tail is doubly exponential and, if $\beta<D/(D-1)$, the tail has a power law. When $\beta=D/(D-1)$, the tail is exponentially distributed.

Asymptotic independence of queues under randomized load balancing

Randomized load balancing greatly improves the sharing of resources while being simple to implement. In one such model, jobs arrive according to a rate-?N Poisson process, with ?<1, in a system of N rate-1 exponential server queues. In Vvedenskaya et al. (Probl. Inf. Transm. 32:15---29, 1996), it was shown that when each arriving job is assigned to the shortest of D, D?2, randomly chosen queues, the equilibrium queue sizes decay doubly exponentially in the limit as N??. This is a substantial improvement over the case D=1, where queue sizes decay exponentially. The reasoning in Vvedenskaya et al. (Probl. Inf. Transm. 32:15---29, 1996) does not easily generalize to jobs with nonexponential service time distributions. A modularized program for treating randomized load balancing problems with general service time distributions was introduced in Bramson et al. (Proc. ACM SIGMETRICS, pp. 275---286, 2010). The program relies on an ansatz that asserts that, for a randomized load balancing scheme in equilibrium, any fixed number of queues become independent of one another as N??. This allows computation of queue size distributions and other performance measures of interest. In this article, we demonstrate the ansatz in several settings. We consider the least loaded balancing problem, where an arriving job is assigned to the queue with the smallest workload. We also consider the more difficult problem, where an arriving job is assigned to the queue with the fewest jobs, and demonstrate the ansatz when the service discipline is FIFO and the service time distribution has a decreasing hazard rate. Last, we show the ansatz always holds for a sufficiently small arrival rate, as long as the service distribution has 2 moments.

Recursive solution of queue length distribution for [formula omitted] queue with single server vacation and variable input rate

In this paper we consider discrete time G e o / G / 1 queue with single server vacation and variable input rate. Using renewal process, probability decomposition technique and u -transform, we derive the recursive expressions of the queue length distributions at epochs n − , n + , and n . The results obtained in this paper indicate that the equilibrium queue length distribution no longer follows the stochastic decomposition discipline. Furthermore we derive the important relations between equilibrium queue length distributions at different epochs ( n − , n + , n ).

Randomized load balancing with general service time distributions

Randomized load balancing greatly improves the sharing of resources in a number of applications while being simple to implement. One model that has been extensively used to study randomized load balancing schemes is the supermarket model. In this model, jobs arrive according to a rate-nλ Poisson process at a bank of n rate-1 exponential server queues. A notable result, due to Vvedenskaya et.al. (1996), showed that when each arriving job is assigned to the shortest of d ≥ 2 randomly chosen queues, the equilibrium queue sizes decay doubly exponentially in the limit as n to ∞. This is a substantial improvement over the case d=1, where queue sizes decay exponentially. The method of analysis used in the above paper and in the subsequent literature applies to jobs with exponential service time distributions and does not easily generalize. It is desirable to study load balancing models with more general, especially heavy-tailed, service time distributions since such service times occur widely in practice. This paper describes a modularized program for treating randomized load balancing problems with general service time distributions and service disciplines. The program relies on an ansatz which asserts that any finite set of queues in a randomized load balancing scheme becomes independent as n to ∞. This allows one to derive queue size distributions and other performance measures of interest. We establish the ansatz when the service discipline is FIFO and the service time distribution has a decreasing hazard rate (this includes heavy-tailed service times). Assuming the ansatz , we also obtain the following results: (i) as n to ∞, the process of job arrivals at any fixed queue tends to a Poisson process whose rate depends on the size of the queue, (ii) when the service discipline at each server is processor sharing or LIFO with preemptive resume, the distribution of the number of jobs is insensitive to the service distribution, and (iii) the tail behavior of the queue-size distribution in terms of the service distribution for the FIFO service discipline.

Limit process of stationary TASEP near the characteristic line

AbstractThe totally asymmetric simple exclusion process (TASEP) on\input amssym ${\Bbb Z}$ with the Bernoulli‐ρ measure as an initial condition, 0 &lt; ρ &lt; 1, is stationary. It is known that along the characteristic line, the current fluctuates at an order of t1/3. The limiting distribution has also been obtained explicitly. In this paper we determine the limiting multipoint distribution of the current fluctuations moving away from the characteristics by the order t2/3. The main tool is the analysis of a related directed last percolation model. We also discuss the process limit in tandem queues in equilibrium. © 2010 Wiley Periodicals, Inc.

Transient Queue Size Distribution Solution of Geom / G / 1 Queue with Feedback-A Recursive Method

This paper considers the Geom / G / 1 queueing model with feedback according to a late arrival system with delayed access (LASDA). Using recursive method, this paper studies the transient property of the queue size from the initial state N(0+) = i. Some new results about the recursive expression of the transient queue size distribution at any epoch n+ and the recursive formulae of the equilibrium distribution are obtained. Furthermore, the recursive formulae of the equilibrium queue size distribution at epoch n−, and n are obtained, too. The important relations between stationary queue size distributions at different epochs are discovered (being different from the relations given in M / G / 1 queueing system). The model discussed in this paper can be widely applied in all kinds of communications and computer network.

Exact waiting time and queue size distributions for equilibrium M/G/1 queues with Pareto service

This paper solves the problem of finding exact formulas for the waiting time cdf and queue length distribution of first-in-first-out M/G/1 queues in equilibrium with Pareto service. The formulas derived are new and are obtained by directly inverting the relevant Pollaczek-Khinchin formula and involve single integrals of non-oscillating real valued functions along the positive real line. Tables of waiting time and queue length probabilities are provided for certain parameter values under heavy traffic conditions.

Stability Analysis of Explicit Congestion Control Protocols

In the context of explicit congestion control protocols like XCP and RCP where the equilibrium queue lengths are zero, we show that the stability region derived from traditional Nyquist analysis is not an accurate representation of the actual stability region, and that the use of switched linear system models with time delay and new Lyapunov tools can provide sound sufficient stability conditions.

Fluid-flow analysis of TCP Westwood with RED

This paper concerns TCP Westwood (TCPW), a recently developed modification of TCP, in combination with RED queue management. We develop a fluid-flow model of TCPW with RED and use it to study both equilibrium and dynamic features. On the equilibrium side, we identify the scaling of window with loss-probability, and compare it to TCP NewReno. We also use the model to find the boundary of stability, beyond which we see large oscillations; we find that the stable region of TCPW is enhanced with respect to TCP NewReno. Furthermore we show preliminary evidence that oscillations, when they occur, have a limited impact on network throughput. Fairness and friendliness of TCPW is also analyzed and compared with NewReno. Our results show that TCPW is more fair than NewReno for flows with heterogenous round trip time. We also found that TCPW friendliness to NewReno is dependent on the difference between equilibrium queue size and the product of bandwidth and delay.

Services within a Busy Period of an M/M/1 Queue and Dyck Paths

We analyze the service times of customers in a stable M/M/1 queue in equilibrium depending on their position in a busy period. We give the law of the service of a customer at the beginning, at the end, or in the middle of the busy period. It enables as a by-product to prove that the process of instants of beginning of services is not Poisson. We then proceed to a more precise analysis. We consider a family of polynomial generating series associated with Dyck paths of length 2n and we show that they provide the correlation function of the successive services in a busy period with n+1 customers.

Response times in M/M/s fork-join networks

We study a fork-join processing network in which jobs arrive according to a Poisson process and each job splits into m tasks, which are simultaneously assigned to m nodes that operate like M/M/s queueing systems. When all of its tasks are finished, the job is completed. The main result is a closed-form formula for approximating the distribution of the network's response time (the time to complete a job) in equilibrium. We also present an analogous approximation for the distribution of the equilibrium queue length (the number of jobs in the system), when each node has one server. Kolmogorov-Smirnov statistical tests show that these formulae are good fits for the distributions obtained from simulations.

Response times in M/M/s fork-join networks

We study a fork-join processing network in which jobs arrive according to a Poisson process and each job splits into m tasks, which are simultaneously assigned to m nodes that operate like M/M/s queueing systems. When all of its tasks are finished, the job is completed. The main result is a closed-form formula for approximating the distribution of the network's response time (the time to complete a job) in equilibrium. We also present an analogous approximation for the distribution of the equilibrium queue length (the number of jobs in the system), when each node has one server. Kolmogorov-Smirnov statistical tests show that these formulae are good fits for the distributions obtained from simulations.

A linear system analysis of RED

The random early detection (RED) algorithm proposed by Floyd and Jacobson [IEEE/ACM Trans. Networking 1 (4) (1993) 397] is a simple and quite effective mechanism for active queue management and as a result has been widely implemented in Internet routers. While RED certainly performs better than drop-tail, its parameters are difficult to tune to give good performance under different congestion scenarios. Also, studies have shown that RED does not stabilize the queue lengths in routers, because its equilibrium queue length strongly depends on the number of active TCP connections. As a result numerous variants of RED have been proposed to work around the performance problems of RED. In this paper, using inferences drawn from a linear systems analysis, we provide an insight into why RED is difficult to tune. The same linear systems analysis is used to show that the DRED algorithm proposed in [Computer 36 (2–3) (2001) 203; Comput. Commun. 24 (12) 2001 1170] overcomes the performance limitations of RED [IEEE/ACM Trans. Network. 1 (4) (1993) 397].

DRED: a random early detection algorithm for TCP/IP networks

AbstractIt is now widely accepted that a RED [2] controlled queue certainly performs better than a drop‐tail queue. But an inherent weakness of RED is that its equilibrium queue length cannot be maintained at a preset value independent of the number of TCP active connections. In addition, RED's optimal parameter setting is largely correlated with the number of connections, the round‐trip time, the buffer space, etc. In light of these observations, we propose DRED, a novel algorithm which uses the basic ideas of feedback control to randomly discard packets with a load‐dependent probability when a buffer in a router gets congested. Over a wide range of load levels, DRED is able to stabilize a router queue occupancy at a level independent of the number of active TCP connections. The benefits of stabilized queues in a network are high resources utilization, predictable maximum delays, more certain buffer provisioning, and traffic‐load‐independent network performance in terms of traffic intensity and number of connections. Copyright © 2002 John Wiley &amp; Sons, Ltd.