Low-priority Queue Research Articles

Performance comparison of several priority schemes with priority jumps

In this paper, we consider several discrete-time priority queues with priority jumps. In a priority scheduling scheme with priority jumps, real-time and non-real-time packets arrive in separate queues, i.e., the high- and low-priority queue respectively. In order to deal with possibly excessive delays however, non-real-time packets in the low-priority queue can in the course of time jump to the high-priority queue. These packets are then treated in the high-priority queue as if they were real-time packets. Many criteria can be used to decide when packets of the low-priority queue jump to the high-priority queue. Some criteria have already been introduced in the literature, and we first overview this literature. Secondly, we propose and analyse a new priority scheme with priority jumps. Finally, we extensively compare all cited schemes. The schemes all differ in their jumping mechanism, based on a certain jumping criterion, and thus all have a different performance. We show the pros and cons of each jumping scheme.

Dynamic QoS queuing control mechanism for multimedia differentiated services

An essential goal of communication networks is to provide multimedia services with QoS streaming. A properly designed multimedia QoS system must reserve requested resources according to user QoS requirements and the available network resources. However, the static resource allocation among priority queues in DiffServ networks leads to insufficient resource usage when a burst occurs in one priority queue while other queues starve. This study presents a User-Oriented QoS Streaming System to achieve perceptible satisfaction based on novel streaming and media differentiation policies in DiffServ networks. This study also proposes that the Dynamic QoS Queue Mapping (DQ 2M) mechanism dynamically control queue scheduling by adaptively maximizing the utilization of queues and network resources according to the soft states of the DiffServ network. Evaluation results indicate that the proposed DQ 2M algorithm can improve the fairness and efficiency of resource utilization for low-priority queues.

A closed form solution of Bernoullian two-classes priority queue

We investigate the effect of a two-priority policy when serving slotted traffic arriving with Bernoulli statistics. We are able to give explicit closed-form expressions for the distribution of the lengths of the high- and low-priority queue.

A closed form solution of bernoullian two-classes priority queue

We investigate the effect of a two-priority policy when serving slotted traffic arriving with Bernoulli statistics. We are able to give explicit closed-form expressions for the distribution of the lengths of the high- and low-priority queue.

Packet Models Revisited: Tandem and Priority Systems

We examine two extensions of traditional single-node packet-scale queueing models: tandem networks and (strict) priority systems. Two generic input processes are considered: periodic and Poisson arrivals. For the two-node tandem, an exact expression is derived for the joint distribution of the total queue length, and the queue length of the first queue, implicitly determining the distribution of the second queue. Similarly we derive the distribution of the low-priority queue in a two-class priority system. We also provide explicit approximations based on the Brownian bridge.

A Dynamic Priority Queue Model for Simultaneous Service of Two Traffic Types

We consider a priority queue with a dynamic, queue-length-threshold scheduling policy. Customers are classed into two types (type-1 and type-2), and the service order of the two classes depends on the queue length of the type-1 queue. The high priority (type-2) class (e.g., voice) is served until the low priority (e.g., data) queue exceeds the threshold L, at which time service is given to the low priority class until its queue length decreases to L. The arrivals of the two classes follow independent Poisson processes, and the service time of each customer has an exponential distribution with parameter $\mu$. We derive the balance equations in the steady state, and explicitly obtain the joint probability generating function for the queue lengths of the two customer classes. This gives the joint queue length distribution as an integral. We then obtain detailed asymptotic results for the joint distribution. In particular, we study the tail behavior. We also discuss heavy traffic diffusion approximations for this model.

Prioritized failure recovery in communication networks and its transient analysis

Prioritized recovery can provide differentiated recovery services to the users in communication networks. This paper studies the transient behavior of communication networks under prioritized recovery. Stochastic Petri net models are constructed to capture the behaviors of a queueing system at four different phases during the recovery process, i.e. failure-free phase, failure is detected, higher priority service is recovered, and lower priority service is recovered. The final state of the stochastic Petri net model in one phase is used as the initial state of the model in the next phase. Both the traffics for the higher and lower priority queue are modeled as Markov Modulated Poisson Processes. Two queues are served with round-robin fashion. With the help of Stochastic Petri Net Package (SPNP), the transient performances of the queueing system during each phase are numerically studied. The queue with the higher recovery priority significantly outperforms the queue with the lower recovery priority during the recovery process. A new concept about performance recovery is introduced. Performance recovery time, which is defined as the time required for a performance index to reach a desired and acceptable value, if it is deteriorated by the service disruption, is distinct from and more important than the failure recovery time. The models and analytical methods presented in this paper can be used to quantify the performance recovery time.

IPv6 traffic with multi-class QoS in VPN

This paper focuses on the modeling and performance analysis for IPv6 traffic with multi-class QoS in virtual private networks (VPNs). The multi-class QoS is implemented on differentiated service basis using priority scheme of 4 bits defined in the packet header of IPv6. A VPN-enabled IP router is modeled as a tandem queuing system in which each output link consists of two parallel priority output queues. The high-priority queue is used to carry the delay sensitive traffic while the low-priority queue is used to carry the delay insensitive traffic. On the other hand, multiple thresholds are implemented in each queue, respectively, for packet loss priority control. The performance analysis is done using fluid flow techniques. The numerical results obtained from the analysis show that the differentiated service based on the priority schemes defined in IPv6 is able to effectively satisfy the multi-class QoS requirement for supporting multimedia services in VPN. The performance trade-off between the delay sensitive traffic and delay insensitive traffic in terms of traffic throughput, packet loss probability and end-to-end delay in VPN networks is presented.

Modeling and performance analysis for IPv6 traffic with multiple QoS classes

This paper focuses on the modeling and performance analysis for IPv6 traffic with multi-class QoS in virtual private networks (VPN). The multi-class QoS is implemented on differentiated service basis using priority scheme of 4 bits defined in the packet header of IPv6. A VPN-enabled IP router is modeled as a tandem queuing system in which each output link consists of two parallel priority output queues. The high-priority queue is used to carry the delay sensitive traffic while the low-priority queue is used to carry the delay insensitive traffic. On the other hand, multiple thresholds are implemented in each queue, respectively, for packet loss priority control. The performance analysis is done using fluid flow techniques. The numerical results obtained from the analysis show that the differentiated service based on the priority schemes defined in IPv6 is able to effectively satisfy the multi-class QoS requirement for supporting multimedia services in VPN. The performance trade-off between the delay sensitive traffic and delay insensitive traffic in terms of traffic throughput, packet loss probability and end-to-end delay in VPN networks is presented.

Correctness of compiling Occam to transputer code

This paper contributes to the development of a rigorous mathematical framework for the study of provably correct compilation techniques. The proposed method is developed through an implementation of a real-life non-toy imperative programming language with non-determinism and parallelism, i.e. Occam, to a commercial machine, namely the Transputer. We provide a mathematical definition of the Transputer Instruction Set architecture for executing Occam together with a correctness proof for a general compilation schema of Occam programs into Transputer code. We start from the ground model, an abstract processor, running a high and a low priority queue of Occam processes, which formalizes the semantics of Occam at the abstraction level of atomic Occam instructions. We develop here increasingly more refined levels of Transputer semantics, proving correctness (and when possible also completeness) for each refinement step. Along the way we collect our proof assumptions, a set of natural conditions for a compiler to be correct, thus making our proof applicable to a large class of compilers. As a by-product our construction provides a challenging realistic case study for proof verification by theorem provers.

Approximate analysis of timer-controlled priority scheme in the single-service token-passing systems

This paper presents an approximate analysis of delay performance in the timer-controlled priority scheme of single-service token passing systems, where each priority queue is allowed to transmit one message at a time. This is the practical case of real-time networks such as those used in aircraft and spacecraft control, process control and factory communications. The token passing systems consist of N stations, each of which has (K+1) priority queues. The highest priority queues can transmit a message whenever the token arrives. A lower priority queue can transmit a message if the token rotation time (which was reset and restarted at the previous token arrival instant) is not expired when the token arrives. An analytical model is developed based on the independent and identically distributed random variables of conditional effective service times (time interval between two consecutive instants at which a priority queue has an opportunity to transmit a message) of each priority queue. The condition is assigned on the basis of the contribution of message transmission time to the effective service time. The approximate mean waiting time for each priority queue is determined using the well-known results from Kuehn (1979), who first introduced the conditional cycle time. The approximate analytical model is validated by comparison with the simulation results.< <ETX xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">&gt;</ETX>

Mean message delays for two packet-FIFO queueing disciplines

Expressions are derived for the mean message delay, as a function of message length, when data messages of different lengths arrive asynchronously at a trunk and are divided, as they come in, into packets of some maximum length. The packets of different messages are intermingled and are put onto the trunk either from a single first-in-first-out (FIFO) queue, or from a high-priority and a low-priority queue for single-packet and multiple-packet messages, respectively. The results are compared to the case in which messages enter service in order of arrival and each message is served to completion without interruption, so that the mean message delay is independent of message length. >

A Useful Tool in the Theory of Priority Queueing

A double input channel priority network, consisting of a low-priority queue with general interarrival distribution and Markovian service time and a high priority completely Markovian queue, is considered. A simple and handy estimate of the mean queue length for the two flows is presented together with its relative effectiveness. The analysis results show that the channel utilization can be improved by introducing a deterministic traffic flow without significant degradation of the high-priority queue performance. These results can be used in designing a communication network, where many packetized traffic flows with different input statistics and delay threshold requirements (packetized voice, video, and data) have to be served by the same switching and transmission facilities.

A Queuing Model of a Time-Sliced Priority-Driven Task Dispatching Algorithm

A queuing analysis of a preemptive, priority-driven, time-sliced dispatcher algorithm, typical of those found in many time-sharing systems, is presented. A distinctive feature of the system being modeled is that a preempted task, when readmitted to the dispatcher, is reallocated a full time slice rather than the residual. The combination of preemption, time-slice renewal and class-dependent arrival and service rates places the analysis beyond that of the well-known FBN algorithms. The model is used to predict the behavior of the dispatcher algorithm under various workloads. The model reveals that, at high arrival rates, the time-slice renewal policy prevents processor bound tasks from being relegated to the lower priority queues, thereby severely degrading the throughput of tasks with short processor requirements.

M/G/1 round robin discipline

This paper analyzes a round robin M/G/1 queuing discipline with quantum sizes that depend on the number of circuits a job has made. Tight bounds on the steady state mean response conditional on job size are found. Results show that the round robin discipline is relatively insensitive to changes in workload and follows the processor sharing response curves surprisingly closely, even for large quantum sizes. The analysis allows determination of a proper cutoff point, the point at which a job would be sent to a lower priority queue. Also derived is the Laplace-Stieltjes transform of the response of a job that arrives to find a given workload. The transient mean response is calculated for this case.

Synthesis of a Feedback Queueing Discipline for Computer Operation

Considerable effort has been invested in devising and analyzing sequencing rules for multiprogrammed or time-shared systems. A much studied discipline of this kind is the so-called system with feedback to lower priority queues. This discipline contains many parameters, in general, which must be fixed in order to achieve the desired waiting-time performance of the discipline. In this paper the problem of synthesizing a system of the above type is solved, by setting parameter values so that prespecified waiting time criteria are satisfied, assuming Poisson arrival and general service time parameters are known.

Two Queues Under Preemptive Priority with Poisson Arrival and Service Rates

This paper extends the work of White and Christie, reported in Operations Research, vol 6, pp 79-95 (January-February, 1958), on preemptive priority queuing. A random walk procedure yields recursive relations between the state probabilities in the steady state. Recursive relations for the moments of the length of the lower priority queue follow. Combined with the use of moment generating functions, this approach provides moments of the waiting time for elements of the lower priority class that arrive when the system is in a given state, i.e., the two queues are of given lengths. The moments of waiting time and time in the system are also obtained recursively for the steady state. Routines for computing and checking are provided.