Fair Bandwidth Research Articles

Shrew Attack Prevention in RED Queue with Partial Flow Analysis

Shrew Attacks or Low Rate Denial of Service(LDoS) Attacks are initiated by sending large amount of packets for very short span of time such that the packet sending rate crosses the link capacity resulting in network congestion. Compared to Denial of Service (DoS) Attack, LDoS attack is very difficult to be detected because, the attacker can maintain low average packet sending rate while executing an attack. If the rate and interval of LDoS attack is properly estimated and executed, this attack can cause a severe threat to the retransmission time out adjustment of TCP and hence reduce its throughput to near zero. This paper proposes a lightweight LDoS filter which can be added with Preferential Dropping RED, to detect and prevent LDoS packets before they reach RED dropping policy. The advantage of this method is that only partial flows need to be analyzed to detect an attack. Simulations done in NS2 shows that, our method can effectively mitigate LDoS attack while maintaining fairness in bandwidth and low average queuing delay. General Terms Networking, Security, Intrusion Detection System

Performance Evaluation of Controlling High Bandwidth Flows by RED-PD

This paper proposed to investigate and evaluate the performance of one of the algorithm used to provide fair bandwidth allocation to the flows. First In First Out (FIFO) queuing is simple but does not protect responsive flows from unresponsive one, flows that are sending more than their fair share. One of FIFO queuing is Random Early Detection which it can effectively avoid congestion at routers, but it also cannot provide fair bandwidth for the flows. On the other hand, per-flow scheduling mechanisms provide max-min fairness but are more complex, it requires keeping state for all flows going through the router; it’s proved that high bandwidth flows at the time of congestion consume most of the bandwidth of the link, so this algorithm (RED-PD) is most candidates to provide fairness to the flows. Simulations with networks demonstrate that there are chances for RED-PD to enhance its work, by means of incorporating the test of unresponsive flows actively in response to changes in the packet drop rate.

Rotating Preference Queues: An Efficient Queue Management Scheme for Fair Bandwidth Allocation

In this study, an efficient queue management scheme is proposed to achieve fair bandwidth sharing by rotating preference queues (RPQ). RPQ consists of a set of first-in-first-out (FIFO) output queues that are dynamically rotated. In addition, RPQ selects accepted packets and dispatches them to adequate output queues depending on packet distribution and preference. Using simulations, we show that RPQ tends to approach DRR, and it significantly outperforms CSFQ, DDE, CHOKe and FIFO in terms of fairness. RPQ has a complexity of O(1) and is simple to implement in high-speed networks.

An Intelligent Adaptive Arbiter for Maximum CPU Utilization, Fair Bandwidth Allocation and Low Latency

Rapid advancement in the world of microprocessor necessitates solving the bus contention in a most efficient manner. The major challenge is to reduce the latency of the system and to achieve a fair bandwidth allocation using maximum CPU utilization. Combination of Adaptive Arbitration (AA) algorithm and the processors according to their traffic behavior can be one of the feasible options to tackle the aforesaid problems due to its fair bandwidth allotment and low system latency. This article provides a comprehensive picture of the global scenario of AA algorithm for fair bandwidth allocation so as to enable researchers to decide the direction of further investigation. Some of the works published so far in this area are reviewed, classified according to their objectives, and presented in an organized manner with general conclusion.

Improving Bandwidth Utilization and Fairness between TCP Flows based on a Machine-learning Approach

Improving Bandwidth Utilization and Fairness between TCP Flows based on a Machine-learning Approach

Fair Bandwidth Sharing in Network Using FCFS-TP Protocol

Fair Bandwidth Sharing in Network Using FCFS-TP Protocol

A Hybrid OFDM-TDM Architecture with Decentralized Dynamic Bandwidth Allocation for PONs

One of the major challenges of passive optical networks is to achieve a fair arbitration mechanism that will prevent possible collisions from occurring at the upstream channel when multiple users attempt to access the common fiber at the same time. Therefore, in this study we mainly focus on fair bandwidth allocation among users, and present a hybrid Orthogonal Frequency Division Multiplexed/Time Division Multiplexed architecture with a dynamic bandwidth allocation scheme that provides satisfying service qualities to the users depending on their varying bandwidth requirements. Unnecessary delays in centralized schemes occurring during bandwidth assignment stage are eliminated by utilizing a decentralized approach. Instead of sending bandwidth demands to the optical line terminal (OLT) which is the only competent authority, each optical network unit (ONU) runs the same bandwidth demand determination algorithm. ONUs inform each other via signaling channel about the status of their queues. This information is fed to the bandwidth determination algorithm which is run by each ONU in a distributed manner. Furthermore, Light Load Penalty, which is a phenomenon in optical communications, is mitigated by limiting the amount of bandwidth that an ONU can demand.

Survey on transport control in data center networks

Traditional fair bandwidth sharing by leveraging AIMD-based congestion control mechanisms faces great challenges in data center networks. Much work has been done to solve one of the various challenges. However, no single transport layer protocol can solve all of them. In this article, we focus on the transport layer in data centers, and present a comprehensive survey of existing problems and their current solutions. We hope that this article can help readers quickly understand the causes of each problem and learn about current research progress, so as to help them make new contributions in this field.

Enhanced DBA to Provide QoS to Coexistent EPON and 10G-EPON Networks

The evolution of the Ethernet passive optical network (EPON) to the enhanced 10G-EPON standard looks poised to maintain its position as the leading fiber to the home technology worldwide. Fair bandwidth allocation will be a challenge in the near future in plants where 1G and 10G customer premises equipment coexists. In this paper we propose a distributed algorithm, distributed dynamic scheduling for EPON (DDSPON), to dynamically allocate the bandwidth requested by the user equipment with quality of service (QoS) constraints. The proposed algorithm is scalable and simple to re-configure, which facilitates a smoother transition from legacy 1G to 10G networks. By using the DDSPON, enhanced optical network units (ONUs) will take full advantage of the 10 Gb/s rate while legacy 1G ONUs maintain their service level agreement; as a result, the individual performance is maximized for both groups of ONUs. We also present an analytical model to evaluate the algorithm performance in a coexistent 1G/10G network. Finally, we also illustrate the enhancement of the DDSPON to provision the QoS to different services under the DiffServ architecture. Analysis and simulation results validate the effectiveness of the proposed mechanisms.

BFTDT: Byzantine Fault Tolerance tryout for Dependable Transactions in Cloud

Cloud Web Services (CWS) is the technology used for business collaboration and integration among the web users. The Web Services Atomic Transactions (WS-AT) have been used for the trusted distributed transaction processing over the web. The WS-AT in the distributed sense has byzantine faults to overcome that Byzantine Faults Techniques (BFT) is used. The reliable coordinator provides the services that are Coordination services, Activation services, Registration Services and Completion services which make the transaction effective and reliable. In the trusted environment, to evade congestion of the resources, fair share bandwidth allocation scheme is used to allocate separate bandwidth for each web users and the transaction is processed Coordinator server and the Transaction Processing Monitor (TPM). The WS-AT for business applications analysis shows the high degree of dependability, security, trust, fault tolerance and fairness of the resources in the trusted environment.

Mobility: A Double-Edged Sword for HSPA Networks: A Large-Scale Test on Hong Kong Mobile HSPA Networks

This paper presents an empirical study on the performance of mobile High Speed Packet Access (a 3.5G cellular standard usually abbreviated as HSPA) networks in Hong Kong via extensive field tests. Our study, from the viewpoint of end users, covers virtually all possible mobile scenarios in urban areas, including subways, trains, off-shore ferries, and city buses. We have confirmed that mobility has largely negative impacts on the performance of HSPA networks, as fast-changing wireless environment causes serious service deterioration or even interruption. Meanwhile, our field experiment results have shown unexpected new findings and thereby exposed new features of the mobile HSPA networks, which contradict commonly held views. We surprisingly find out that mobility can improve fairness of bandwidth sharing among users and traffic flows. Also, the triggering and final results of handoffs in mobile HSPA networks are unpredictable and often inappropriate, thus calling for fast reacting fallover mechanisms. Moreover, we find that throughput performance does not monotonically decrease with increased mobility level. We have conducted in-depth research to furnish detailed analysis and explanations to what we have observed. We conclude that mobility is a double-edged sword for HSPA networks. To the best of our knowledge, this is the first public report on a large-scale empirical study on the performance of commercial mobile HSPA networks.

FairTorrent: A Deficit-Based Distributed Algorithm to Ensure Fairness in Peer-to-Peer Systems

Peer-to-peer file-sharing applications suffer from a fundamental problem of unfairness. Free-riders cause slower download times for others by contributing little or no upload bandwidth while consuming much download bandwidth. Previous attempts to address this fair bandwidth allocation problem suffer from slow peer discovery, inaccurate predictions of neighboring peers' bandwidth allocations, underutilization of bandwidth, and complex parameter tuning. We present FairTorrent, a new deficit-based distributed algorithm that accurately rewards peers in accordance with their contribution. A FairTorrent peer simply uploads the next data block to a peer to whom it owes the most data as measured by a deficit counter. FairTorrent is resilient to exploitation by free-riders and strategic peers, is simple to implement, requires no bandwidth overallocation, no prediction of peers' rates, no centralized control, and no parameter tuning. We implemented FairTorrent in a BitTorrent client without modifications to the BitTorrent protocol and evaluated its performance against other widely used BitTorrent clients. Our results show that FairTorrent provides up to two orders of magnitude better fairness, up to five times better download times for contributing peers, and 60%-100% better performance on average in live BitTorrent swarms.

Auction-Based Bandwidth Allocation in Multi-Hop Wireless Ad Hoc Networks

In this paper, we propose a multi-hop auction-based bandwidth allocation mechanism to address the flow contention problem in wireless ad hoc networks. By modeling the problem as an iterative auction-based structure, it enables us to derive fair and efficient bandwidth allocation to each node on the basis of only local information. Further, a multi-hop flow coordination mechanism is then developed to optimize the network performance. Simulation results suggest that the proposed mechanism outperforms other approaches in terms of network throughput, bandwidth utilization, fairness, end-to-end delay, packet loss rate, and robustness.

A game-based algorithm for fair bandwidth allocation in Fibre-Wireless access networks

Fibre-Wireless (FiWi) access networks have been proposed as flexible and cost-effective solutions for future access networks. At the wireless mesh section, wireless routers have to forward both local traffic from directly connected users and foreign traffic from neighbour wireless routers. How to allocate resources to local and foreign traffic at each router in a balanced way, while avoiding starvation of routers requiring less resources, is a fundamental issue that must be solved so that new services emerge. Here, we develop a repeated game framework for bandwidth allocation and propose an algorithm that allocates bandwidth in a fair manner. The algorithm is able to detect over claiming routers and avoid possible denial of service that these may cause to others. Moreover, unfruitful use of resource is prevented, avoiding the forwarding of packets that would be dropped at some point later in the path, and queueing delay conditions are kept similar among local and foreign traffic. These fair network conditions open way for QoS support since it is easier to ensure the operationality of services.

An aggregate marker for bandwidth fairness in DiffServ

Surveys and studies on DiffServ exhibit existence of unfairness between aggregates in the excess bandwidth region. Many researchers have proved that providing a proportional fair share of the bandwidth between aggregates solve this. Based on this, aggregate markers such as ItswTCM and I2tswTCM are found to improve bandwidth fairness in DiffServ network among aggregates, when the aggregates are responsive in nature. However, co-existence of non-responsive aggregates with responsive aggregates leads to unfairness. Hence, in this paper an attempt has been made to evade the above unfairness by a new TSW based marker; where the drop precedence of the packet is arrived at an adaptive manner based on its estimated rate and the availability of the network resource. The unique feature of this proposed algorithm is that it provides enhanced fairness not only among responsive aggregates but also between responsive and non-responsive aggregates. Further, simulations have been carried out in NS-2 with changes in RTT, target rate of the aggregates and the number of flows in the aggregate. The results show that the proposed marker reduces the influence of these factors on the fairness performance exhibiting the suitability of the marker for different network topologies.

A Fair and Secure Bandwidth Allocation for AMI Mesh Network in Smart Grid

Great interest and much research have emerged to implement and improve the smart grid. Even though the communication network realizes or improves end-to-end smartness at every part of the smart grid, the advanced metering infrastructure (AMI) enhances the smartness more effectively by enabling a two-way communication between customers and the Control Center. We have many possible solutions for the underlying networks for the AMI system, and yet, wireless mesh networks (WMNs) draw much attention in the community because of their flexibility and robustness. One of the most important issues to consider is how to solve the well-known bottleneck problem of WMNs. In this paper, we propose the E-fee scheme to securely protect the collaboration among nodes in the AMI system when WMNs are employed as the underlying network of the system. Then we conduct a theoretical analysis on the protection capability and we also see the results of a performance evaluation study to verify if the E-fee scheme can address the problem. The results indicate that the E-fee scheme is a secure and fair bandwidth allocation technique for the AMI system based on WMSs.

Congestion in large balanced multirate networks

In this paper, we obtain analytical approximations for various performance measures for a large fluid stochastic network that operates under a balanced fair bandwidth allocation policy. Balanced fairness results in the insensitivity of the stationary distribution of the number in the system to the precise distribution of file sizes. Balanced fairness has been shown to coincide with proportional fairness in large systems. The model we consider is that of servers operating under balanced fair rate allocations that are accessed by a large number of independent heterogeneous flows characterized by their arrival rate and general distributions of the file sizes; and a maximum service rate associated with each type of flow. The largeness of the system is parameterized by a scaling parameter that scales the arrival rates and capacity in such a way that the ratio is fixed. By exploiting a connection of the congestion probabilities with multirate Erlang loss systems, we use local limit large deviation methods to obtain accurate approximations as the scaling increases. The paper first discusses the single link case which is then extended to the case of a parking lot model that is a special case of tree networks.

Efficient and Fair Bandwidth Allocation in Multichannel Cognitive Radio Networks

Cognitive radio (CR) improves spectrum efficiency by allowing secondary users (SUs) to dynamically exploit the idle spectrum owned by primary users (PUs). This paper studies optimal bandwidth allocation of SUs for throughput efficiency. Consider the following tradeoff: an SU increases its instantaneous throughput by accessing more spectrum, but channel access/switching overhead, contention among multiple SUs, and dynamic PU activity create higher liability for larger bandwidths. So how much is too much? In this paper, we study the optimal bandwidth allocation for multiple SUs. Our approach is twofold. We first study the optimal bandwidth an SU should use to maximize the per-SU throughput in the long term. The optimal bandwidth is derived in the context of dynamic PU activity, where we consider both independent and correlated PU channel scenarios while accounting for the effects of channel switching overhead. We further consider the case of suboptimal spectrum use by SUs in the short term due to PU activity dynamics. We propose an efficient channel reconfiguration (CREC) scheme to improve SUs' performance. We use real PU channel activity traces in the simulations to validate our results. The work sheds light on the design of spectrum sharing protocols in cognitive radio networks.

Hardware implementation study of several new egress link scheduling algorithms

The provision of Quality of Service (QoS) in interconnection networks is required for new multimedia and time-sensitive applications, which are very important for recent utility computing data centers (UCDCs) using high performance networks. These interconnection networks support switch-based principles and establish high demands in terms of bandwidth, time-delay, and delivery over short distances. A key component for networks with QoS support is the egress link scheduling algorithm. Apart from providing a good performance in terms of, for example, good end-to-end delay (also called latency) and fair bandwidth allocation, an ideal scheduling algorithm implemented in a high-performance network with QoS support should satisfy another important property which is to have a low computational and implementation complexity. In this paper, we propose specific implementations (taking into account the characteristics of current high performance networks) of several fair-queuing scheduling algorithms and compare their complexity in terms of silicon area and computation delay. In order to carry out this comparison, we have devised our own hardware comparison methodology. Following this methodology, we have performed our own hardware implementation for the different schedulers. We have modeled the schedulers using the Handel-C language and employed the DK design suite tool from Celoxica in order to obtain hardware estimates on silicon area and arbitration time.

Design a novel fairness model in WiMAX mesh networks

WiMAX mesh is acting as a promising alternative to construct high speed and wide range multi-hop wireless access network; however the resource scheduling with respect to user fairness and system efficiency in such networks is still remained unspecified. To perform the global fair bandwidth allocation for all competing traffic flows, we design a new fairness model called multihop max–min fair (MMF) to be applied in WiMAX mesh network. The MMF fairness is firstly designed based on the homogenous scenario where every link holds the same transmission capacity; then it is extended to the heterogeneous scenario where different links have different transmission capacity. As for the scheduling for all those competing traffic flows over the mesh network, we first get the global fair share of scheduling opportunities for each flow according to the MMF principle, then determine the resource allocation for each flow and each traffic link. The relevant resource scheduling algorithm is given with an analysis on various performance metrics as well. The simulation results validate the superiority of our proposed scheme compared with the existing schemes.