A queueing model for a storage buffer in which the arrival rate is controlled by a switch with a random delay

doi:10.1016/0166-5316(85)90025-2

Performance Evaluation

Volume 5, Issue 4, November 1985, Pages 243-256

https://doi.org/10.1016/0166-5316(85)90025-2 Get rights and content

Abstract

In order to control overload conditions in a message storage buffer, the arrival rate is often reduced when the content of the buffer exceeds a level K and is allowed to return to its previous value when the buffer content drops below a level L < K. In a number of practical situations, the action of the switch is not instantaneous and a period of random duration elapses before the desired lower arrival rate is attained. During that period, the arrival rate is only partially reduced or not at all. This paper deals with a descriptive queueing model for a buffer of this type and also proposes algorithms useful in its design. Various extensions, which may be analyzed by means of the same methodlogy, are briefly described.

References (11)

C.M. Harris
Queues with stochastic service rates
Nav. Res. Logist. Quart.
(1967)
C.M. Harris
A queueing system with multiple service time distributions
Nav. Res. Logist. Quart.
(1967)
D.M. Lucantoni et al.
Numerical methods for a class of Markov chains arising in queueing theory
D.M. Lucantoni
A GI/M/c queue with a different service rate for customers who need not wait — An algorithmic solution
Cah. Centre d'Études Rech. Oper.
(1982)
M.F. Neuts
Moment formulas for the Markov renewal branching process
Adv. Appl. Prob.
(1976)

There are more references available in the full text version of this article.

Cited by (8)

Performance analysis of the DAR(1)/D/c priority queue under partial buffer sharing policy
2004, Computers and Operations Research
We analyze a multi-server priority queueing system with partial buffer sharing, where the input is a discrete autoregressive process of order 1 (DAR(1)) which is known as a good mathematical model for a VBR-coded teleconference video traffic. We assume that arriving packets are classified into two priority classes, say, high priority class and low priority class based on their importance. A threshold T is set to limit the accessibility of low priority packets to the buffer. When the partial buffer sharing is applied to the real time traffic such as teleconference video traffic, it is known that it can decrease the queueing delay at the expense of the loss of low priority packets which is less important. Since the queueing delay is more important than the loss probability for real time traffic, it is important to analyze the queueing delay of DAR(1) arrivals under the partial buffer sharing policy. Based on the Wiener–Hopf factorization of the GI/GI/1 queue, we obtain the waiting time distribution of a packet which is not discarded at its arrival in the steady state. Numerical examples are provided to show the feasibility of our analysis. We also show that the partial buffer sharing policy significantly decreases the waiting time as the value of threshold increases.
Overload effects and their prevention
1991, Performance Evaluation
The so-called “thrashing effect”, well known from virtual storage, but also reported from data-base systems and packet switching networks, has turned out to be a common phenomenon of large systems with concurrent processing. It simply means that beyond a saturation point an increase of the load (e.g. number of jobs) leads to a (sometimes sudden) decrease in performance (e.g. throughput). With growing size and complexity of computer systems and the general trend towards distribution, overload phenomena of different origin can interfere and superimpose mutually, resulting in a composite overload effect that can hardly be broken down into its constituents. Because the complexity of such systems defies detailed modeling, it is more appropriate to look at those systems in a more macroscopic, behavioral way, considering only the two externally measurable variables “load” and “throughput”. The resulting abstraction from internal details can smooth the way to a more general treatment and application. The article deals with such overload phenomena and their prevention in a general way using a control-theoretic approach. Special emphasis is placed on dynamic behavior, where load characteristics are changing with time, making feedback mechanisms necessary. The problem is approached as a dynamic optimum search problem for which different algorithms are presented and compared by simulation.
A hysteretic source rate control scheme for a finite buffer in a wireless environment
2007, IEEE International Conference on Communications
Martingale methods for analysing single-server queues
2002, Queueing Systems
Analysis of a Rate-Based Access Control Mechanism for High-speed Networks
1993, IEEE Transactions on Communications
Priority Management in ATM Switching Nodes
1991, IEEE Journal on Selected Areas in Communications

View all citing articles on Scopus

^☆: This research was supported by the National Science Foundation under Grant No. ECS-8205404.

View full text

A queueing model for a storage buffer in which the arrival rate is controlled by a switch with a random delay☆

Abstract

Queues with stochastic service rates

Nav. Res. Logist. Quart.

A queueing system with multiple service time distributions

Nav. Res. Logist. Quart.

Numerical methods for a class of Markov chains arising in queueing theory

A GI/M/c queue with a different service rate for customers who need not wait — An algorithmic solution

Cah. Centre d'Études Rech. Oper.

Moment formulas for the Markov renewal branching process

Adv. Appl. Prob.