Queue Balancing Research Articles

A finite state Markovian queue to let in impatient customers only during K-vacations

We investigate a matrix analysis study for a single-server Markovian queue with finite capacity, i.e. an M/M/1/N queue, where the single server can go for a maximum, i.e. a K number of consecutive vacation periods. During these vacation periods of the server, every customer becomes impatient and leaves the queues. If the server detects that the system is idle during service startup, the server rests. If the vacation server finds a customer after the vacation ends, the server immediately returns to serve the customer. Otherwise, the server takes consecutive vacations until the server takes a maximum number of vacation periods, e.g. K, after which the server is idle and waits to serve the next arrival. During vacation, customers often lose patience and opt for scheduled deadlines independently. If the customer’s service is not terminated before the customer’s timer expires, the customer is removed from the queue and will not return. Matrix analysis provides a computational form for a balanced queue length distribution and several other performance metrics. We design a ‘no-loss; no-profit cost model’ to determine the appropriate value for the maximum value of K consecutive vacation periods and provide a solution with a numerical illustration.

Experimental Survey of FPGA-Based Monolithic Switches and a Novel Queue Balancer

This paper studies small to medium-sized monolithic switches for FPGA implementation and presents a novel switch design that achieves high algorithmic performance and FPGA implementation efficiency. Crossbar switches based on virtual output queues (VOQs) and variations have been rather popular for implementing switches on FPGAs, with applications in network switches, memory interconnects, network-on-chip (NoC) routers etc. The implementation efficiency of crossbar-based switches is well-documented on ASICs, though we show that their disadvantages can outweigh their advantages on FPGAs. One of the most important challenges in such input-queued switches is the requirement for iterative scheduling algorithms. In contrast to ASICs, this is more harmful on FPGAs, as the reduced operating frequency and narrower packets cannot “hide” multiple iterations of scheduling that are required to achieve a modest scheduling performance.Our proposed design uses an output-queued switch internally for simplifying scheduling, and a queue balancing technique to avoid queue fragmentation and reduce the need for memory-sharing VOQs. Its implementation approaches the scheduling performance of a state-of-the-art FPGA-based switch, while requiring considerably fewer resources.

Study of improved signal‐based merge strategy in work zone areas based on Cellular Automata simulation

Work zones have critical effects on traffic safety and efficiency. Previous researchers have developed lots of merge strategies to regulate drivers’ behaviours, which can be divided into early merge and late merge. As a special kind of late merge, signal-based merge strategy was proposed in recent years to reduce traffic conflicts and increase throughput at bottlenecks. Although traffic signal has been confirmed to improve the traffic in work zone areas effectively, the existing strategy still has limitations that require further improvements. In this research, an improved signal-based merge (IM) strategy was proposed to fulfil the research gaps. The strategy included a calculation method of green time considering the relationship between arrival traffic and the bottleneck's capacity and a dynamic signal-phase conversion method. Based on the Cellular Automata simulation, the effectiveness of the IM strategy was verified through parameter sensitive analysis and comparative study with the existing merge strategies. The results showed that the IM strategy could lead to greater throughputs, lower traffic delays, less aggressive merge and more balanced queue length. Furthermore, the IM strategy also showed great stability in response to the changes of parameters and the imbalance of arrival traffic. Overall, the proposed IM strategy is considered more effective in improving the traffic management in work zone areas and is feasible to be applied in practice.

Optimizing distribution of metered traffic flow in perimeter control: Queue and delay balancing approaches

Perimeter traffic flow control based on the macroscopic or network fundamental diagram provides the opportunity of operating an urban traffic network at its capacity. Because perimeter control operates on the basis of restricting inflow via reduced green times at selected entry (gated) links, vehicles on those links may be subject to queuing and delay. The experienced delay or resulting queue lengths depend on the adopted policy for the distribution of the inflows and corresponding green times at the gated links. The chosen policy may have a significant impact on the traffic system under control. For example, managing queue lengths may reduce the interference with upstream traffic whereas the management of delays may improve users’ perception with respect to equity and fairness. In this paper, an approach has been proposed to distribute the gated flow based on the queue lengths or experienced delay at the gated signalized junctions. This is in contrast to standard practice that distributes inflows proportionally to the gated links’ saturation flows. Perimeter control is then evaluated in a microscopic simulator for a realistic traffic network and compared in three configurations against fixed-time: perimeter control without queue or delay management; perimeter control with relative queue balancing; and perimeter control with delay balancing. It has been found that managing the queues at the gated links not only improves the overall network performance but also reduces the possibility of queue propagation to the upstream junctions. This improves traffic flow outside the protected network by managing the queue propagation at the gated links and reducing the possibility of queue spill-back to upstream intersections. In addition, the results indicate that perimeter control with delay balancing has a similar performance as the case without queue or delay management being a suitable approach for flow distribution among the gated links. In the scenarios with perimeter control with either queue or delay balancing the gap between the ordered flow by the controller and the actual flow crossing the stop-line at the gated links reduced remarkably.

Model Predictive Control of a Road Junction

This paper presents a model predictive control (MPC) approach for optimally managing the traffic light (TL) signals at a road junction. The objective is to improve queue balancing compared to traditional control strategies where TL signals are periodic. The resulting MPC optimization problem is of quadratic mixed-integer nature. The proposed approach is validated via simulations based on a real scenario.

Excel в помощь аналитику: формируем сбалансированную очередь распределения ресурсов в системе

Excel в помощь аналитику: формируем сбалансированную очередь распределения ресурсов в системе

Improvement and research of the method of balanced queue management on interfaces of telecommunication network routers

In the work the improvement of a method for balanced queue management on the interface of a telecommunication network router is proposed. The novelty of the proposed method is to provide a consistent solution to the problems of aggregation and distribution of packet flows among queues (Congestion Management tasks), as well as problems of balancing the allocation of the interface band-width among the class queues (Resource Allocation Tasks) in accordance with the principles of the Traffic Engineering Queues concept. Unlike the known approaches, the proposed solution has a positive effect on the computational complexity of the calculations, which is connected, first of all, with the decrease in the size of optimization problems and their linear nature. This was conditioned by the fact that the consistency of the Congestion Management and Resource Allocation processes was ensured on the basis of their agreed solution, since the decision on aggregation and distribution of flows among queues was not connected with the results of the solutions of the Resource Allocation problem. The study of Congestion Management and Resource Allocation processes using the proposed method has confirmed its efficiency and adequacy of the obtained solutions. The aggregation and distribution of packet flows in corresponding queues were carried out in accordance with the values of their classes. While the optimal balanced allocation of the interface bandwidth among the formed class queues was carried out in accordance with the principles of the concept of Traffic Engi-neering Queues - the queue with a higher value of the class had a lower utilization factor.

Mathematical model for searching the optimal resources size for the virtual service node

In the article the work of the on-line charging system namely the functioning of the Diameter-server is investigated. Considered the possibility of expanding the system of service by attracting additional computational resources organized according to the cloud technology. The purpose of the study is to calculate the optimal parameters of the service server, namely computing resources such as operational and disk memory, processor time, resource of the input and output system, to service the billing request flows at a given level of quality, such QoS parameters as control of losses and delays in maintenance. Conclusions . The mathematical model of the task of choosing the power of the servicing device, where maintenance is carried out at a given level of quality, is proposed. Unlike the existing load balancing approaches, when additional maintenance devices are added between the queue balancing, to provide increased load, the proposed model allows for the scalability of the resources of service nodes built on cloud computing technology to be taken into account. The proposed model allows to calculate the parameters of the Diameter-server for servicing the input stream and to provide a reduction of the average delay in service of the request and minimize the losses associated with the overload server service.

Traffic model for the interconnection of networks and operators using MPLS-TE

In this paper, the main features of MPLS Traffic Engineering are presented to illustrate how telecommunication service providers use them to create interconnections between each other in order to offer telecom services satisfying QoS commitments. Based on previous traffic models, a new model, which deals with traffic queue balancing for different Classes of Service, and for a provider using another provider´s network is presented. The model output shows that carrying another operator’s traffic may increase delays in an undesirable manner, forcing the carrier to increase the serving rate of LSRs until Utilization is below 60 %. In order to validate the model, a number of network scenarios are implemented in the Wolfram Mathematica 10.1 Study Version, based on study case configurations of an MPLS network. The total global model is useful for future implementation of test-beds of interconnected providers under an MPLS environment.

BeaQoS: Load balancing and deadline management of queues in an OpenFlow SDN switch

Current OpenFlow specification is unable to set the service rate of the queues inside OpenFlow devices. This lack does not allow to apply most algorithms for the satisfaction of Quality of Service requirements to new and established flows. In this paper we propose an alternative solution implemented through some modifications of Beacon, one popular SDN controller. It acts as follows: using ‘almost’-real-time statistics from OpenFlow devices, Beacon will re-route flows on different queues to guarantee the observance of deadline requirements (e.g. the flow is still useful if, and only if, is completely received by a given time) and/or an efficient queue balancing in an OpenFlow SDN switch. Differently from the literature, we do not propose any new primitive or modification of the OpenFlow standard: our mechanism, implemented in the controller, works with regular OpenFlow devices. Our changes in the SDN controller will be the base for the design of a class of new re-routing algorithms able to guarantee deadline constraints and queue balancing without any modification of the OpenFlow specification, as well as, of OpenFlow devices.

Stochastic programming model for oversaturated intersection signal timing

In large cities, signalized intersections often become oversaturated in rush hours due to growing traffic demand. If not controlled properly, they may collectively result in serious congestion. How to schedule traffic signals for oversaturated intersections has thus received increasing interests in recent years. Among various factors that may influence control performance, uncertainty in traffic demand remains as an important one that needs to be further studied. In some recent works, e.g., Yin (2008) and Li (2011), robust optimization models have been utilized to address uncertainty in traffic demand and to design fixed-timed signal control for oversaturated intersections. In this paper, we propose a stochastic programming (SP) model to schedule adaptive signal timing plans that minimize the expected vehicle delay. Our numerical experiments show that the proposed SP model better describes the fluctuations of traffic flows and outperforms the deterministic linear programming (LP) model in total vehicle delay, total throughput, and vehicle queue lengths. Moreover, we compare the proposed SP model with the adaptive signal control model proposed in Lin et al. (2011) to provide insights on such improvements from green time utilization and queue balancing perspectives. Furthermore, we study the feasibility of the proposed SP model in practice, with an emphasis on the required sample sizes.

Decentralized Queue Balancing and Differentiated Service Scheme Based on Cooperative Control Concept

In this paper, we introduce the concept of a bottleneck-routers cooperation in the explicit rate-control framework of communication networks in order to mitigate congestion effects on the network performance and balance the queues. The proposed controller at each router (server or switch) regulates the rates of the heterogeneous source classes leveraging on the cooperation of neighboring bottlenecks. We consider the model of multibottleneck network in the presence of time delay and formulate global stability conditions suitable for network parameters and controller gains design. The proposed approach guarantees good performance in terms of link utilization, packet loss and fairness. Additionally it is guaranteed queue balancing without requiring rerouting or hop-by-hop operation differently from the existing approaches. A validation is carried out by a discrete packet experiment simulator in a realistic multibottleneck scenario to demonstrate the effectiveness of the key idea of the paper. Finally the proposed scheme is compared to some of well-known network controller-type presented in the literature in both steady-state and dynamic network scenario.

Queue balancing of load and expedition service in a cement industry in Brazil

The load and weight process in a cement industry is one of logistic step that shows the biggest time of occurrence, increasing the queues. This study aims to do a scenarios to solve this queue problem. This way, it pretends to find an better resources distribuition.

A consensus based rate control scheme for ATM networks

This paper is concerned with the design and validation of a consensus based rate control for the Asynchronous Transfer Mode (ATM) networks. A sufficient condition for network closed loop stability in the presence of multiple bottleneck and heterogeneous sources with different time delays is given and it is used for switch-controller parameters design. The stability condition and the resulting congestion control performance in terms of robustness to network uncertainties, fairness, link utilization and packet loss are validated by discrete packet simulator. Moreover, it ensured convergence of the queue length to the desired consensus steady-state value with resulting network queue balancing.

On the efficacy of fine-grained traffic splitting protocols in data center networks

Current multipath routing techniques split traffic at a per-flow level because, according to conventional wisdom, forwarding packets of a TCP flow along different paths leads to packet reordering which is detrimental to TCP. In this paper, we revisit this "myth" in the context of cloud data center networks which have regular topologies such as multi-rooted trees. We argue that due to the symmetry in the multiple equal-cost paths in such networks, simply spraying packets of a given flow among all equal-cost paths, leads to balanced queues across multiple paths, and consequently little packet reordering. Using a testbed comprising of NetFPGA switches, we show how cloud applications benefit from better network utilization in data centers.

Distributed Resource Allocation Based on Queue Balancing in Multihop Cognitive Radio Networks

Cognitive radio (CR) allows unlicensed users to access the licensed spectrum opportunistically (i.e., when the spectrum is left unused by the licensed users) to enhance the spectrum utilization efficiency. In this paper, the problem of allocating resources (channels and transmission power) in multihop CR networks is modeled as a multicommodity flow problem with the dynamic link capacity resulting from dynamic resource allocation, which is in sharp contrast with existing flow-control approaches that assume fixed link capacity. Based on queue-balancing network flow control that is ideally suited for handling dynamically changing spectrum availability in CR networks, we propose a distributed scheme (installed and operational in each node) for optimal resource allocation without exchanging spectrum dynamics information between remote nodes. Considering the power masks, each node makes resource-allocation decisions based on current or past local information from neighboring nodes to satisfy the throughput requirement of each flow. Parameters of these proposed schemes are configured to maintain the network stability. The performance of the proposed scheme for both asynchronous and synchronous scenarios is analyzed comparatively. Both cases of sufficient and insufficient network capacity are considered.

Writeback-aware partitioning and replacement for last-level caches in phase change main memory systems

Phase-Change Memory (PCM) has emerged as a promising low-power main memory candidate to replace DRAM. The main problems of PCM are that writes are much slower and more power hungry than reads, write bandwidth is much lower than read bandwidth, and limited write endurance. Adding an extra layer of cache, which is logically the last-level cache (LLC), can mitigate the drawbacks of PCM. However, writebacks from the LLC might (a) overwhelm the limited PCM write bandwidth and stall the application, (b) shorten lifetime, and (c) increase energy consumption. Cache partitioning and replacement schemes are important to achieve high throughput for multi-core systems. However, we noted that no existing partitioning and replacement policy takes into account the writeback information. This paper proposes two writeback-aware schemes to manage the LLC for PCM main memory systems. Writeback-aware Cache Partitioning (WCP) is a runtime mechanism that partitions a shared LLC among multiple applications. Unlike past partitioning schemes, our scheme considers the reduction in cache misses as well as writebacks. Write Queue Balancing (WQB) replacement policy manages the cache partition of each application intelligently so that the writebacks are distributed evenly among PCM write queues. In this way, applications rarely stall due to unbalanced PCM write traffic among write queues. Our evaluation shows that WCP and WQB result in, on average, 21% improvement in throughput, 49% reduction in PCM writes, and 14% reduction in energy over a state-of-the-art cache partitioning scheme.

On the efficacy of fine-grained traffic splitting protocolsin data center networks

Multi-rooted tree topologies are commonly used to construct high-bandwidth data center network fabrics. In these networks, switches typically rely on equal-cost multipath (ECMP) routing techniques to split traffic across multiple paths, such that packets within a flow traverse the same end-to-end path. Unfortunately, since ECMP splits traffic based on flow-granularity, it can cause load imbalance across paths resulting in poor utilization of network resources. More fine-grained traffic splitting techniques are typically not preferred because they can cause packet reordering that can, according to conventional wisdom, lead to severe TCP throughput degradation. In this work, we revisit this fact in the context of regular data center topologies such as fat-tree architectures. We argue that packet-level traffic splitting, where packets of a flow are sprayed through all available paths, would lead to a better load-balanced network, which in turn leads to significantly more balanced queues and much higher throughput compared to ECMP.

Trading Rate for Balanced Queue Lengths for Network Delay Minimization

We consider a communication channel with two transmitters and one receiver, with an underlying rate region which is approximated as a general pentagon. Different from the Gaussian multiple access channel (MAC) capacity region, the sum-rate on the dominant face of this pentagon is not a constant. We allocate rates from this rate region to users according to their current queue lengths in order to minimize the average delay in the system. We formulate the problem as a Markov decision problem (MDP), and derive the structural properties of the corresponding discounted-cost MDP. We show that the delay-optimal policy has a switch curve structure. For the discounted-cost problem, we prove that the switch curve has a limit along one of the dimensions. The delay-optimal policy divides the entire queue state space into two via a switch curve. If the queue state is on one side of the switch curve, the system operates at one of the corner points of the rate pentagon which favors maximum sum-rate. When the queue state switches to the other side of the switch curve, the system operates at the other corner point of the rate pentagon which favors balancing the queue lengths. As a result, the system does not always operate at the sum-rate maximizing rate pair, but trades rate for balanced queue lengths for the goal of minimizing the overall delay. The existence of a limit in the switch curve along one of dimensions implies that, once the queue state is beyond the limit, the system always operates at one of the corner points, implying that the queues can be operated partially distributedly.

Delay-Optimal Opportunistic Scheduling and Approximations: The Log Rule

This paper considers the design of multiuser opportunistic packet schedulers for users sharing a time-varying wireless channel from performance and robustness points of view. For a simplified model falling in the classical Markov decision process framework, we numerically compute and characterize mean-delay-optimal scheduling policies. The computed policies exhibit radial sum-rate monotonicity: As users' queues grow linearly, the scheduler allocates service in a manner that deemphasizes the balancing of unequal queues in favor of maximizing current system throughput (being opportunistic). This is in sharp contrast to previously proposed throughput-optimal policies, e.g., Exp rule and MaxWeight (with any positive exponent of queue length). In order to meet performance and robustness objectives, we propose a new class of policies, called the Log rule, that are radial sum-rate monotone (RSM) and provably throughput-optimal. In fact, it can also be shown that an RSM policy minimizes the asymptotic probability of sum-queue overflow. We use extensive simulations to explore various possible design objectives for opportunistic schedulers. When users see heterogenous channels, we find that emphasizing queue balancing, e.g., Exp rule and MaxWeight, may excessively compromise the overall delay. Finally, we discuss approaches to implement the proposed policies for scheduling and resource allocation in OFDMA-based multichannel systems.