Showing papers on "Queue management system published in 2013"

PIE: A lightweight control scheme to address the bufferbloat problem

Abstract: Bufferbloat is a phenomenon where excess buffers in the network cause high latency and jitter. As more and more interactive applications (e.g. voice over IP, real time video conferencing and financial transactions) run in the Internet, high latency and jitter degrade application performance. There is a pressing need to design intelligent queue management schemes that can control latency and jitter; and hence provide desirable quality of service to users. We present here a lightweight design, PIE (Proportional Integral controller Enhanced), that can effectively control the average queueing latency to a reference value. The design does not require per-packet extra processing, so it incurs very small overhead and is simple to implement in both hardware and software. In addition, the design parameters are self-tuning, and hence PIE is robust and optimized for various network scenarios. Simulation results, theoretical analysis and Linux testbed results show that PIE can ensure low latency and achieve high link utilization under various congestion situations.

Modelling delay propagation within an airport network

Abstract: We describe an analytical queuing and network decomposition model developed to study the complex phenomenon of the propagation of delays within a large network of major airports. The Approximate Network Delays (AND) model computes the delays due to local congestion at individual airports and captures the " ripple effect" that leads to the propagation of these delays. The model operates by iterating between its two main components: a queuing engine (QE) that computes delays at individual airports and a delay propagation algorithm (DPA) that updates flight schedules and demand rates at all the airports in the model in response to the local delays computed by the QE. The QE is a stochastic and dynamic queuing model that treats each airport in the network as a M(t)/. Ek(t)/1 queuing system. The AND model is very fast computationally, thus making possible the exploration at a macroscopic level of the impacts of a large number of scenarios and policy alternatives on system-wide delays. It has been applied to a network consisting of the 34 busiest airports in the continental United States and provides insights into the interactions through which delays propagate through the network and the often-counterintuitive consequences. Delay propagation tends to " smoothen" daily airport demand profiles and push more demands into late evening hours. Such phenomena are especially evident at hub airports, where some flights may benefit considerably (by experiencing reduced delays) from the changes that occur in the scheduled demand profile as a result of delays and delay propagation. © 2011 Elsevier Ltd.

Measuring the Effect of Queues on Customer Purchases

Abstract: We conduct an empirical study to analyze how waiting in queue in the context of a retail store affects customers' purchasing behavior. Our methodology combines a novel data set with periodic information about the queuing system (collected via video recognition technology) with point-of-sales data. We find that waiting in queue has a nonlinear impact on purchase incidence and that customers appear to focus mostly on the length of the queue, without adjusting enough for the speed at which the line moves. An implication of this finding is that pooling multiple queues into a single queue may increase the length of the queue observed by customers and thereby lead to lower revenues. We also find that customers' sensitivity to waiting is heterogeneous and negatively correlated with price sensitivity, which has important implications for pricing in a multiproduct category subject to congestion effects. This paper was accepted by Martin Lariviere, operations management.

Management of electric vehicle charging station queues

Abstract: There is disclosed method of management of electric vehicle charging station (EVCS) queues and an EVCS queue management system. The method includes limiting access to the EVCS to an individual at the top of an EVCS queue during a changeover time. The changeover time may be predetermined or may be generated based upon data pertaining to the management of the queue and/or to the vehicle of the user at the top of the queue. If the user at the top of the queue fails to activate the EVCS during the changeover time, the next in line in the queue will be given a changeover time in which to reach the EVCS. If the queue is otherwise empty, the user at the top of the queue will be notified that the EVCS is no longer reserved, and that anyone may use the EVCS.

No silver bullet: extending SDN to the data plane

Abstract: The data plane is in a continuous state of flux. Every few months, researchers publish the design of a new high-performance queueing or scheduling scheme that runs inside the network fabric. Many such schemes have been queen for a day, only to be surpassed soon after as methods --- or evaluation metrics --- evolve.The lesson, in our view: there will never be a conclusive victor to govern queue management and scheduling inside network hardware. We provide quantitative evidence by demonstrating bidirectional cyclic preferences among three popular contemporary AQM and scheduling configurations.We argue that the way forward requires carefully extending Software-Defined Networking to control the fast-path scheduling and queueing behavior of a switch. To this end, we propose adding a small FPGA to switches. We have synthesized, placed, and routed hardware implementations of CoDel and RED. These schemes require only a few thousand FPGA "slices" to run at 10 Gbps or more --- a minuscule fraction of current low-end FPGAs --- demonstrating the feasibility and economy of our approach.

Forward request queuing in a distributed edge processing environment

Abstract: A server in a distributed environment includes a process that manages incoming client requests and selectively forwards service requests to other servers in the network. The server includes storage in which at least one forwarding queue is established. The server includes code for aggregating service requests in the forwarding queue and then selectively releasing the requests, or some of them, to another server. The queuing mechanism preferably is managed by metadata, which, for example, controls how many service requests may be placed in the queue, how long a given service request may remain in the queue, what action to take in response to a client request if the forwarding queue's capacity is reached, etc. In one embodiment, the server generates an estimate of a current load on an origin server (to which it is sending forwarding requests) and instantiates the forward request queuing when that current load is reached.

Adaptive Duty Cycle Control with Queue Management in Wireless Sensor Networks

Abstract: This paper proposes a control-based approach to the duty cycle adaptation for wireless sensor networks. The proposed method controls the duty cycle through the queue management to achieve high-performance under variable traffic rates. To have energy efficiency while minimizing the delay, we design a feedback controller, which adapts the sleep time to the traffic change dynamically by constraining the queue length at a predetermined value. In addition, we propose an efficient synchronization scheme using an active pattern, which represents the active time slot schedule for synchronization among sensor nodes, without affecting neighboring schedules. Based on the control theory, we analyze the adaptation behavior of the proposed controller and demonstrate system stability. The simulation results show that the proposed method outperforms existing schemes by achieving more power savings while minimizing the delay.

Queue management system and method

Abstract: A queue management system in accordance with present embodiments may include a data server system including a processor and memory. The queue management system may include an interface system of the data server system configured to receive a reservation request including an indication of a general time period of arrival of a guest to an area. Additionally, the queue management system may include a reservation right allotment system of the data server system configured to correlate a reservation slot for accessing an attraction to identification information for the guest. Further, the queue management system may include a detection system of the data server system configured to determine when a portable identification feature associated with the identification information has arrived to the area, and a reservation assignment system of the data server system configured to establish a reservation to access the attraction at a specific time window.

A Skiplist-Based Concurrent Priority Queue with Minimal Memory Contention

Abstract: Priority queues are fundamental to many multiprocessor applications. Several priority queue algorithms based on skiplists have been proposed, as skiplists allow concurrent accesses to different parts of the data structure in a simple way. However, for priority queues on multiprocessors, an inherent bottleneck is the operation that deletes the minimal element. We present a linearizable, lock-free, concurrent priority queue algorithm, based on skiplists, which minimizes the contention for shared memory that is caused by the DeleteMin operation. The main idea is to minimize the number of global updates to shared memory that are performed in one DeleteMin. In comparison with other skiplist-based priority queue algorithms, our algorithm achieves a 30 - 80% improvement.

Modelling network flow with and without link interactions: the cases of point queue, spatial queue and cell transmission model

Abstract: This paper studies dynamic network models in which congestion takes the form of queuing behind bottlenecks. The three models in consideration, the point queue (P-Q) model, the spatial queue (S-Q) model and the cell transmission model, all deal with queues caused by temporal bottlenecks, yet differ with each other in describing the S-Q characteristics. Our work focuses on exploring the impacts of spatial distribution of queues on the result of dynamic network loading (DNL). For this purpose, simple node models (merge and diverge) are introduced to describe flow interactions across links. We present a numerical solution algorithm to the DNL problem which accommodates the three models in a general framework. Analytical solutions are derived for simple cases and agree with numerical results. Our limited experiments indicate that P-Q and S-Q may severely underestimate network travel costs under DNL, particularly when queue spillback prevails.

Traffic and load aware dynamic queue management

Abstract: Some embodiments provide a queue management system that efficiently and dynamically manages multiple queues that process traffic to and from multiple virtual machines (VMs) executing on a host This system manages the queues by (1) breaking up the queues into different priority pools with the higher priority pools reserved for particular types of traffic or VM (eg, traffic for VMs that need low latency), (2) dynamically adjusting the number of queues in each pool (ie, dynamically adjusting the size of the pools), (3) dynamically reassigning a VM to a new queue based on one or more optimization criteria (eg, criteria relating to the underutilization or overutilization of the queue)

Measuring human queues using WiFi signals

Abstract: We investigate using smartphone WiFi signals to track human queues, which are common in many business areas such as retail stores, airports, and theme parks. Real-time monitoring of such queues would enable a wealth of new applications, such as bottleneck analysis, shift assignments, and dynamic workflow scheduling. We take a minimum infrastructure approach and thus utilize a single monitor placed close to the service area along with transmitting phones. Our strategy extracts unique features embedded in the signal traces to infer the critical time points when a person reaches the head of the queue and finishes service, and from these inferences we derive a person's waiting and service times. We develop a feature driven approach in our system. Extensive experiments conducted both in the laboratory demonstrate that our system is robust to queues with different waiting time. We show that in spite of noisy signal readings, our methods can measure important time periods in queue (e.g., service and waiting times) to within a $10$ second resolution.

NADA: A Unified Congestion Control Scheme for Low-Latency Interactive Video

Abstract: Low-latency, interactive media applications (e.g., video conferencing) present a unique set of challenges for congestion control. Unlike TCP, the transport mechanism for interactive media needs to adapt fast to abrupt changes in available bandwidth, accommodate sluggish responses and output rate fluctuations of a live video encoder, and avoid high queuing delay over the network. An ideal scheme should also make effective use of all types of congestion signals from the network, including packet losses, queuing delay, and explicit congestion notification (ECN) markings. This paper presents a unified approach for congestion control of interactive video: network-assisted dynamic adaptation (NADA). In NADA, the sender regulates its sending rate based on a composite congestion signal calculated and reported by the receiver, which combines both implicit (e.g., loss and delay) and explicit (e.g., ECN marking) congestion indications from the network. Via a consistent set of sender adaptation rules, the scheme can reap the full benefits of proactive, explicit congestion notifications supported by advanced queue management schemes. It remains equally responsive in the absence of such notifications. Extensive simulation studies show that NADA interact well with a wide variety of queue management schemes: conventional drop-tail, random early detection (RED), recently proposed CoDel (controlled delay) and PIE (Proportional Integral controller Enhanced), as well as a token-bucket-based random marking scheme based on Pre-Congestion Notification (PCN). Furthermore, NADA reacts fast to changes over the network, allows for weighted bandwidth sharing among multiple competing video streams, and sustains a substantial share of bottleneck bandwidth when coexisting with TCP.

A Discrete Event Simulation Model for Evaluating the Performances of an M/G/C/C State Dependent Queuing System

Abstract: M/G/C/C state dependent queuing networks consider service rates as a function of the number of residing entities (e.g., pedestrians, vehicles, and products). However, modeling such dynamic rates is not supported in modern Discrete Simulation System (DES) software. We designed an approach to cater this limitation and used it to construct the M/G/C/C state-dependent queuing model in Arena software. Using the model, we have evaluated and analyzed the impacts of various arrival rates to the throughput, the blocking probability, the expected service time and the expected number of entities in a complex network topology. Results indicated that there is a range of arrival rates for each network where the simulation results fluctuate drastically across replications and this causes the simulation results and analytical results exhibit discrepancies. Detail results that show how tally the simulation results and the analytical results in both abstract and graphical forms and some scientific justifications for these have been documented and discussed.

Using mixture of Gamma distributions for Bayesian analysis in an M/G/1 queue with optional second service

Abstract: The paper proposes Bayesian framework in an M/G/1 queuing system with optional second service. The semi-parametric model based on a finite mixture of Gamma distributions is considered to approximate both the general service and re-service times densities in this queuing system. A Bayesian procedure based on birth-death MCMC methodology is proposed to estimate system parameters, predictive densities and some performance measures related to this queuing system such as stationary system size and waiting time. The approach is illustrated with several numerical examples based on various simulation studies.

Adaptive active queue management controller for TCP communication networks using PSO-RBF models

Abstract: Addressing performance degradations in end-to-end congestion control has been one of the most active research areas in the last decade. Active queue management (AQM) is a promising technique to congestion control for reducing packet loss and improving network utilization in transmission control protocol (TCP)/Internet protocol (IP) networks. AQM policies are those policies of router queue management that allow for the detection of network congestion, the notification of such occurrences to the hosts, and the adoption of a suitable control policy. Radial bias function (RBF)-based AQM controller is proposed in this paper. RBF as a nonlinear controller is suitable as an AQM scheme to control congestion in TCP communication networks since it has nonlinear behavior. Particle swarm optimization (PSO) algorithm is also employed to derive RBF output weights such that the integrated-absolute error is minimized. Furthermore, in order to improve the robustness of RBF controller, an error-integral term is added to RBF equation. The output weights and the coefficient of the integral error term in the latter controller are also optimized by PSO algorithm. It should be noted that in both proposed controllers the parameters of radial basis functions are selected to symmetrically partition the input space. The results of the comparison with adaptive random early detection (ARED), random exponential marking (REM), and proportional-integral (PI) controllers are presented. Integral-RBF has better performance not only in comparison with RBF but also with ARED, REM and PI controllers in the case of link utilization while packet loss rate is small.

On the Effectiveness of CoDel for Active Queue Management

Abstract: Internet in the present scenario has become a gigantic source of information. There has been a tremendous rise in the variety of Internet applications, with each application demanding a specific performance criteria to be satisfied. Routers presently use Passive Queue Management (PQM) mechanisms and hence, merely have any control over the queue occupancy. Therefore, there has been an increased interest in exploring Active Queue Management (AQM) in Internet routers so as to reduce the queue latency and meet the demands of time sensitive applications. In this paper, we mainly focus on analyzing the effectiveness of a recently proposed AQM mechanism called Controlled Delay (CoDel). We study the effectiveness of CoDel by carrying out simulations in ns-2 and comparing its performance with existing AQM mechanisms in variety of Internet scenarios. Based on the simulation results obtained, we discuss the advantages and shortcomings of CoDel in terms of bottleneck link utilization, mean queue length and packet drop rate.

Linear random access queue

Abstract: Methods and systems for managing queues of sample carriers in a bidirectional track allow queues to behave as linear FIFO queues while providing random access to samples by resorting samples in the queue. One or more storage positions to allow sample carriers to be selectively removed and reinserted into the queue, allowing the order of sample carriers to be resorted.

On Applying Cuckoo Search Algorithm to Positioning GI/M/1/N Finite-Buffer Queue with a Single Vacation Policy

Abstract: In the article is presented possible application of swarm intelligence in the positioning of computer or queuing systems. As an object recognition for swarm intelligence is given a problem of cost optimization of the GI/M/1/N-type queue with finite buffer and a single vacation policy. Positioning of the analyzed queuing system was done using swarm intelligence, in particular cuckoo search algorithm (CS), basing on the explicit representation for the joint transform of the first busy period, first idle time and the number of packets transmitted during the first busy period and fixed values of unit costs of the server's functioning during the service. An optimal set of system parameters is found by CS for exponentially distributed vacation period and 2-Erlang distribution of inter arrival times.

RED-FT: A Scalable Random Early Detection Scheme with Flow Trust against DoS Attacks

Abstract: In traditional Active Queue Management algorithms, e.g. RED, each flow, defined with the source and destination IP address of packets, fairly contends for the cache of bottleneck queues. However a malicious flow cannot be identified. And it enables potential network-layer attacks, e.g. the flooding Denial-of-Service (DoS) attack and the low-rate DoS attack. In this letter, we propose a new scheme using the flows trust values to defend against DoS attacks. Different from previous schemes, it employs the flow trust to safeguard legitimate flows. A router monitors network flows and calculates flows trust values, which are used for the relevant queue management. Malicious flows would be with lower trust values while legitimate flows would be with higher ones. Simulation results show that the scheme improves the throughput and delay in DoS attacking scenarios comparing with existing queue management algorithms. We consider the scheme is practical and effective to secure networks.

Queue management based duty cycle control for end-to-end delay guarantees in wireless sensor networks

Abstract: In this paper, we propose an analytical method for duty cycle adaptation in wireless sensor networks so that delay requirement is guaranteed while power consumption is minimized. The proposed method, named Dual-QCon, provides a formal method for stabilizing controller design based on queue management in order to control both duty cycle and queue threshold according to changing network conditions. Dual-QCon also provides a delay notification mechanism in order to determine an appropriate queue threshold of each node according to the application-dependent and time-varying delay requirements. Based on control theory, we analyze the adaptive behavior of the proposed method and derive conditions for system stability. Asymptotic analysis shows that Dual-QCon guarantees end-to-end delay requirement by controlling parameters of local nodes. Simulation results indicate that Dual-QCon outperforms existing scheduling protocols in terms of delay and power consumption.

Applications of Queuing Theory for Open-Pit Truck/Shovel Haulage Systems

Abstract: Surface mining is the most common mining method worldwide, and open pit mining accounts for more than 60% of all surface output. Haulage costs account for as much as 60% of the total operating cost for these types of mines, so it is desirable to maintain an efficient haulage system. As the size of the haulage fleet being used increases, shovel productivity increases and truck productivity decreases, so an effective fleet size must be chosen that will effectively utilize all pieces of equipment. One method of fleet selection involves the application of queuing theory to the haul cycle. Queuing theory was developed to model systems that provide service for randomly arising demands and predict the behavior of such systems. A queuing system is one in which customers arrive for service, wait for service if it is not immediately available, and move on to the next server or exit the system once they have been serviced. Most mining haul routes consist of four main components: loading, loaded hauling, dumping, and unloaded hauling to return to the loader. These components can be modeled together as servers in one cyclic queuing network, or independently as individual service channels. Data from a large open pit gold mine are analyzed and applied to a multichannel queuing model representative of the loading process of the haul cycle. The outputs of the model are compared against the actual truck data to evaluate the validity of the queuing model developed.

Queue resource reallocation strategy for traffic systems in scale-free network

Abstract: In real communication systems, each node has a finite queue length to store packets due to physical constraints. In this paper, we propose a queue resource allocation strategy for traffic dynamics in scale-free networks. With a finite resource of queue, the allocation of queue length on node i is based on Bi, where Bi is the generalized betweenness centrality of node i. The overall traffic capacity of a network system can be evaluated by the critical packet generating rate (Rc). Through the use of the proposed queue allocation scheme for the shortest path protocol and efficient routing protocol, our strategy performs better than the uniform queue length allocation strategy, which is demonstrated by a larger value of the critical generating rate. We also give a method to estimate the network traffic capacity theoretically.

Congestion detection and management at congestion-tree roots

Abstract: A device for performing congestion detection and management at a node of a network may include a congestion management module to monitor an arrival and a departure rate of data packets associated with a queue, a queue size, or a rate of change of the queue size. The congestion management module may identify the queue as a congested queue by one of determining that the arrival rate of the data packets associated with the queue is larger than the departure rate of the data packets associated with the queue, the queue size is larger than a first threshold, or the rate of change of the queue size is larger than a second threshold. The congestion management module may identify the congested queue as a congestion root by determining that the congested queue is not affected by a flow-control signal. A queue buffer temporarily stores data packets associated with the queue.

System and method for determining a cause of network congestion

Abstract: A method and apparatus of a device that determines a cause and effect of congestion in this device is described. In an exemplary embodiment, the device measures a queue group occupancy of a queue group for a port in the device, where the queue group stores a plurality of packets to be communicated through that port. In addition, the device determines if the measurement indicates a potential congestion of the queue group, where the congestion prevents a packet from being communicated within a time period. If potential congestion exists on that queue group, the device further gathers information regarding packets to be transmitted through that port. For example, the device can gather statistics packets that are stored in the queue group and/or new enqueue packets.