Path Load Balancing Research Articles

Measuring congestion-induced performance imbalance in Internet load balancing at scale

Internet load balancers typically use congestion-oblivious hashing algorithms to distribute traffic across load-balanced (LB) paths, resulting in imbalanced load distribution. When congestion occurs, LB paths could experience different levels of congestion, called congestion imbalance. Congestion imbalance has been extensively studied in datacenters (DCs). It is, however, still under-explored in the Internet. In this paper, we take a first step towards measuring congestion imbalance among Internet LB paths at scale. We present Congi, our lightweight prober, that leverages support vector machine (SVM) classifiers to efficiently detect congestion imbalance using latency samples. Our experiments show that Congi is capable of detecting congestion imbalance between uncongested and congested LB paths: on average, the uncongested path has 3x greater throughput and 0.4% lower packet loss rate than the congested one. To measure congestion imbalance at scale, we then use Congi to conduct measurement campaigns from worldwide DCs to millions of/24s. We find that most DCs experience significant congestion imbalance to 36%–43% of end hosts. Lastly, we use Congi to direct web page downloads under congestion imbalance from a campus network, and show that the download time can be reduced by 44% on average for Alexa top sites for clients experiencing network congestion under congestion imbalance.

AOEHO: A New Hybrid Data Replication Method in Fog Computing for IoT Application.

Recently, the concept of the internet of things and its services has emerged with cloud computing. Cloud computing is a modern technology for dealing with big data to perform specified operations. The cloud addresses the problem of selecting and placing iterations across nodes in fog computing. Previous studies focused on original swarm intelligent and mathematical models; thus, we proposed a novel hybrid method based on two modern metaheuristic algorithms. This paper combined the Aquila Optimizer (AO) algorithm with the elephant herding optimization (EHO) for solving dynamic data replication problems in the fog computing environment. In the proposed method, we present a set of objectives that determine data transmission paths, choose the least cost path, reduce network bottlenecks, bandwidth, balance, and speed data transfer rates between nodes in cloud computing. A hybrid method, AOEHO, addresses the optimal and least expensive path, determines the best replication via cloud computing, and determines optimal nodes to select and place data replication near users. Moreover, we developed a multi-objective optimization based on the proposed AOEHO to decrease the bandwidth and enhance load balancing and cloud throughput. The proposed method is evaluated based on data replication using seven criteria. These criteria are data replication access, distance, costs, availability, SBER, popularity, and the Floyd algorithm. The experimental results show the superiority of the proposed AOEHO strategy performance over other algorithms, such as bandwidth, distance, load balancing, data transmission, and least cost path.

An IoT and machine learning‐based routing protocol for reconfigurable engineering application

With new telecommunications engineering applications, the cognitive radio (CR) network-based internet of things (IoT) resolves the bandwidth problem and spectrum problem. However, the CR-IoT routing method sometimes presents issues in terms of road finding, spectrum resource diversity and mobility. This study presents an upgradable cross-layer routing protocol based on CR-IoT to improve routing efficiency and optimize data transmission in a reconfigurable network. In this context, the system is developing a distributed controller which is designed with multiple activities, including load balancing, neighbourhood sensing and machine-learning path construction. The proposed approach is based on network traffic and load and various other network metrics including energy efficiency, network capacity and interference, on an average of 2 bps/Hz/W. The trials are carried out with conventional models, demonstrating the residual energy and resource scalability and robustness of the reconfigurable CR-IoT.

Using UE-VBS for dynamic virtual small cells deployment and backhauling in 5G Ultra-Dense networks

This paper builds on the UE-based Virtual Base Station (UE-VBS) concept, which enables the Smartphones (i.e., UEs) of the general population to be enhanced with Base Station and Relay Node functionalities and be embedded as an integral part of the Mobile Network Infrastructure, offering relief in stressful or overloaded situations. This paper focuses on the selection process of eligible-UEs that will be activated on-the-fly into UE-VBSs to complement and support the Mobile Network infrastructure, either as virtual small cells (VSCs) or virtual relay nodes (VRNs). We adopt a Modified Affinity Propagation Clustering (MAPC) algorithm to select UE-VBSs to act as VSCs for capacity/data rate expansion in areas where the infrastructure is weak and a more effective and agile network operation is needed. Additionally, to provide for a more spectrum efficient backhauling of mobile data traffic from the VSCs towards the BS, the Multi-hop Load balanced Geographical Path selection (MLGP) algorithm is proposed. The MAPC and MLGP algorithms are implemented and evaluated in MATLAB and also validated using the NS3 network simulator, with mobile data traffic modeled as a non-uniform distribution with respect to the time and space domain. The results demonstrated that the proposed approaches improve the network performance in terms of throughput, delay and jitter.

DOLPHIN: Dynamically Optimized and Load Balanced Path for Inter-Domain SDN Communication

Software-Defined Networking has become an integral technology for large scale networks that require dynamic flow management. It separates the control function from data plane devices and centralizes it in a domain controller. However, only a limited number of switches can be managed by a single and centralized controller which introduces challenges such as scalability, reliability, and availability. Distributed controller architecture resolves these issues but also introduces new challenges of uneven load and traffic management across domains. As real-world networks have redundant links, hence a significant challenge is to distribute traffic flows on multiple paths, within a domain, and across multiple independent domains. The selection of ingress and egress switches becomes even more problematic if the intermediate domain is non-cooperative. In this work, we propose a Dynamically Optimized and Load-balanced Path for Inter-domain (DOLPHIN) communication system, a customized solution for different SDN controllers. It provides control beyond the virtual switch elements in intra and inter-domain communication and extends the range of programmability to wireless devices, such as the Internet of Things or vehicular networks. Extensive simulation results show that the traffic load is distributed evenly on multiple links connecting different domains. We model data center communication and 5G vehicular network communication to show that, by load balancing the flow completion times of the different types of network traffic can be significantly improved.

Latency Imbalance Among Internet Load-Balanced Paths

Load balancers choose among load-balanced paths to distribute traffic as if it makes no difference using one path or another. This work shows that the latency difference between load-balanced paths (called latency imbalance), previously deemed insignificant, is now prevalent from the perspective of the cloud and affects various latency-sensitive applications. In this work, we present the first large-scale measurement study of latency imbalance from a cloudcentric view. Using public cloud around the globe, we measure latency imbalance both between data centers (DCs) in the cloud and from the cloud to the public Internet. Our key findings include that 1) Amazon's and Alibaba's clouds together have latency difference between load-balanced paths larger than 20ms to 21% of public IPv4 addresses; 2) Google's secret in having lower latency imbalance than other clouds is to use its own well-balanced private WANs to transit traffic close to the destinations and that 3) latency imbalance is also prevalent between DCs in the cloud, where 8 pairs of DCs are found to have load-balanced paths with latency difference larger than 40ms. We further evaluate the impact of latency imbalance on three applications (i.e., NTP, delay-based geolocation and VoIP) and propose potential solutions to improve application performance. Our experiments show that all three applications can benefit from considering latency imbalance, where the accuracy of delay-based geolocation can be greatly improved by simply changing how ping measures the minimum path latency.

Latency Imbalance Among Internet Load-Balanced Paths

Load balancers choose among load-balanced paths to distribute traffic as if it makes no difference using one path or another. This work shows that the latency difference between load-balanced paths (called latency imbalance ), previously deemed insignificant, is now prevalent from the perspective of the cloud and affects various latency-sensitive applications. In this work, we present the first large-scale measurement study of latency imbalance from a cloud-centric view. Using public cloud around the globe, we measure latency imbalance both between data centers (DCs) in the cloud and from the cloud to the public Internet. Our key findings include that 1) Amazon's and Alibaba's clouds together have latency difference between load-balanced paths larger than 20ms to 21% of public IPv4 addresses; 2) Google's secret in having lower latency imbalance than other clouds is to use its own well-balanced private WANs to transit traffic close to the destinations and that 3) latency imbalance is also prevalent between DCs in the cloud, where 8 pairs of DCs are found to have load-balanced paths with latency difference larger than 40ms. We further evaluate the impact of latency imbalance on three applications (i.e., NTP, delay-based geolocation and VoIP) and propose potential solutions to improve application performance. Our experiments show that all three applications can benefit from considering latency imbalance, where the accuracy of delay-based geolocation can be greatly improved by simply changing how \textttping measures the minimum path latency.

Enhanced cluster-based stable path for load balancing and reducing overhead using artificial bee colony in MANETs

Enhanced cluster-based stable path for load balancing and reducing overhead using artificial bee colony in MANETs

Dynamic Path Planning Algorithms With Load Balancing Based on Data Prediction for Smart Transportation Systems

In modern transportation, traffic congestion has become an urgent problem in large and medium-sized cities. In smart transportation systems, it is an effective solution to design load balancing path planning algorithms that can dynamically adapt to traffic conditions in order to avoid congestion. In this work, a traffic path planning algorithm based on data prediction (TPPDP) is proposed to find the path with the shortest travel time, which is built on a predictive model based on historical traffic data and current traffic information. Furthermore, a path planning algorithm based on data prediction with load balancing (TPPDP-LB) is also proposed, which combines the predicted information and the number of concurrent requests to achieve the path with shortest travel time while maintaining global load balancing. A specific distributed computing framework for TPPDP-LB algorithm is designed to reduce the runtime of the algorithm. The simulation results proved that both TPPDP and TPPDP-LB algorithms have the advantage of shortest travel time, and TPPDP-LB algorithm achieves load balancing of computing. It is also proved that the distributed computing framework designed for TPPDP-LP algorithm can effectively reduce the runtime of system as well as keep the accuracy of algorithm.

An exact and polynomial approach for a bi-objective integer programming problem regarding network flow routing

This paper presents a bi-objective network flow routing problem regarding network load balancing and flow path length. The problem is formulated within the integer programming framework and an exact and polynomial approach based on the ϵ-constraint technique is presented. In each iteration, our algorithm solves a single-objective linear integer programming subproblem. The goal of our approach is to obtain a minimal complete set of Pareto-optimal solutions. Our proposal was evaluated through several computational experiments, which included grid and random networks. The random topology was generated by the Barabási–Albert model and the settings of the network flows defined here are those commonly used in wireless sensor networks and wireless mesh networks. The analysis of the computational results provide a decision maker with valuable information about which factors most affect the solutions, for example, the smallest and largest bottleneck, the size of increment in the shortest path of the last Pareto-optimal solution and the difference between the path lengths of the first and last Pareto-optimal solutions generated.

Enhanced cluster-based stable path for load balancing and reducing overhead using artificial bee colony in MANETS

A routing overhead plays a vital role in mobile ad-hoc networks. The main issue focused on the previous algorithm is a stable path without considering the overhead of route discovery. The path fails in a route and again reinitiates the discovery process which may lead to additional overhead, increase delay and waste of resource utilisation. The proposed new algorithm reduces the routing overhead for cluster based initial route discovery and data transmission using artificial bee colony (CBRTABC) with ad-hoc on demand distance vector (AODV). This algorithm uses honey bees to calculate the neighbour node for route discovery and a path is established. As a result the cluster head maintains a long period of processing time and produces reliable path. Whenever the optimal path is retained then the additional route discovery overhead is reduced. This projected algorithm is computed in the NS2 simulator produced by improving the packet delivery ratio and it reduces the delay and routing overhead.

A fine-grained multi-source measurement platform correlating routing transitions with packet losses

In this paper, we are interested in the relationship between packet losses and routing changes in an operational network. To do so we designed and deployed DCART, a monitoring platform over RENATER, the French research and education network. Our platform collects four data sources using both active and passive measurements in order to unveil their temporal correlations. Active probing allows especially for measuring packet losses on specifically crafted data flows. Those flows explore several load balanced paths and ease the revelation of forwarding loops. Passive monitoring is achieved by listening to all routing updates from IS-IS, the intra-domain routing protocol in use, and by retrieving tickets generated by the Network Operations Center (NOC).During our monitoring campaign, we observe that most of the series of loss were correlated to routing events either because routing changes lead to inconsistent state transitions, or because faulty – and so lossy – links trigger numerous periods of link flapping. In particular, we show that losses due to forwarding loops resulting from inconsistent routing states are quite common when links come back after an outage. We also show that link flapping sometimes induce very long lasting lossy periods frequently unnoticed by the NOC. A lightweight monitoring platform such as DCART could be used to better anticipate recurrent network outages and to improve the ticketing system.

Multiple QoS Parameters-Based Routing for Civil Aeronautical Ad Hoc Networks

Aeronautical ad hoc network (AANET) can be applied as in-flight communication systems to allow aircraft to communicate with the ground, in complement to other existing communication systems to support Internet of Things. However, the unique features of civil AANETs present a great challenge to provide efficient and reliable data delivery in such environments. In this paper, we propose a multiple quality of service parameters-based routing protocol (MQSPR), to improve the overall network performance for communication between aircraft and the ground. The proposed MQSPR integrates path availability period, residual path load capacity and path latency in route selection and presents a broadcast optimization scheme to minimize flooding. The primary aim of MQSPR is to maintain long link durations, achieve path load balancing and reduce end-to-end delay to satisfy the requirements of civil aviation communication services. The simulation scenario and real-world scenario are set up, respectively, and the experimental results show that the proposed MQSPR can achieve high ground connectivity while effectively increase the path durations, improve the packet delivery ratio and perform the path load balancing. In addition, the flexibility of MQSPR is demonstrated by considering weighting factors of path selection parameters.

Robust and Agile System against Fault and Anomaly Traffic in Software Defined Networks

The main advantage of software defined networking (SDN) is that it allows intelligent control and management of networking though programmability in real time. It enables efficient utilization of network resources through traffic engineering, and offers potential attack defense methods when abnormalities arise. However, previous studies have only identified individual solutions for respective problems, instead of finding a more global solution in real time that is capable of addressing multiple situations in network status. To cover diverse network conditions, this paper presents a comprehensive reactive system for simultaneously monitoring failures, anomalies, and attacks for high availability and reliability. We design three main modules in the SDN controller for a robust and agile defense (RAD) system against network anomalies: a traffic analyzer, a traffic engineer, and a rule manager. RAD provides reactive flow rule generation to control traffic while detecting network failures, anomalies, high traffic volume (elephant flows), and attacks. The traffic analyzer identifies elephant flows, traffic anomalies, and attacks based on attack signatures and network monitoring. The traffic engineer module measures network utilization and delay in order to determine the best path for multi-dimensional routing and load balancing under any circumstances. Finally, the rule manager generates and installs a flow rule for the selected best path to control traffic. We implement the proposed RAD system based on Floodlight, an open source project for the SDN controller. We evaluate our system using simulation with and without the aforementioned RAD modules. Experimental results show that our approach is both practical and feasible, and can successfully augment an existing SDN controller in terms of agility, robustness, and efficiency, even in the face of link failures, attacks, and elephant flows.

Load balanced routing scheme for MANETs with power and delay optimisation

Innovative research on mobile ad hoc networks (MANETs) and their wide applications have created a revolution of today's wireless communication. Efficient routing for dynamically changing network and unstable wireless medium is a challenging task. The conventional AODV routing protocol works on selecting the hop count-based shortest path while delivering the packet towards the destination. This often creates uneven load distribution among network nodes. As a result, any central node hopelessly carries heavier loads that significantly deplete node energy with higher packet loss and delay in processing. In this work, a protocol based on power and delay named as power and delay optimised AODV (PDO-AODV) has been presented which is based on load balanced routing strategy in AODV that improves the quality of service for MANETs. The proposed approach takes neighbouring node's power and delay into consideration for selecting load balanced path. Simulation results manifest that the proposed PDO-AODV achieves improved performance as compared to existing AODV routing protocol in terms of packet drop, network delay, throughput, and retransmission attempts.

HyperFatTree: A Large-Scale Tree-Based Network with Low-Radix Switches

To lower the large-scale network cost and energy consumption, we proposed a hierarchical topology with low-radix switches. The hierarchical topology HyperFatTree is designed by combining Fat Tree topology and complete graph topology. Theory analysis of topology characteristic shows that the proposed topology can achieve high cost performance and high scalability. According to the characteristic of HyperFatTree, we designed a minimum path routing algorithm and a load balanced non-minimum path random routing algorithm for the WorstCase traffic. Evaluation shows that the saturated throughput of HyperFatTree is nearly 68.5 % at the scale of 83,232 nodes under Uniform Random traffic, which is higher than both Fat Tree and Dragonfly. It is also important that the throughput reduces only 5 % while the scale has increased 80 times, which shows good smooth scalability. Evaluation on energy efficiency shows that HyperFatTree performs better than Fat Tree and can achieve 3---7 times higher energy efficiency than Dragonfly.

Energy Competent Routing Protocol Design in MANET with Real Time Application Provision

With the increase of mobile activities in technological environment, it is a challenge to develop and maintain robust networking system with provision of high connectivity. In this context, Mobile Adhoc Network (MANET) is a feasible solution because it is dynamic in nature and the self-configured nodes of MANET can communicate intelligently in adverse and dynamic mobile situations. Though Adhoc on Demand Distance Vector (AODV) Routing protocol handles the networking issues efficiently, still power and delay efficiency has always been a challenging task in MANET due to the resource constrained nodes which are supplied with limited battery power during the operation in the network. Hence while performing various networking activities; gradually the battery power reduces resulting in delay in processing of packet and forwarding them to the next node during transmission. This causes overall end to end delay in the total transmission. Therefore many researches are still going on in the direction of designing robust and reliable protocols. In this paper the authors have developed a robust and energy efficient routing protocol for MANET with real time support. Their approach is different as it calculates the remaining residual battery power, bounded delay and packet processing rate of the intermediate node before selecting a node to forward the packet in the direction of destination. The cost function calculated using the above three factors along with the condition of congested node is also checked to correctly choose a load balanced path from source to destination. Simulation results show that the proposed protocol performs better than the conventional AODV protocol in terms of better network life time, throughput, delay and packet delivery ratio.

Network Traffic Load Balancing In Gateways

Load balancing is the practice of evenly distributing work among multiple devices. This technique provides several important benefits. Load Balancing in Gateways is a very important in today's emerging world. All the traffic of the computer network flows through the gateways of its network. A network gateway is an internetworking system capable of joining together two networks that use different base protocols. A network gateway can be implemented completely in software, completely in hardware, or as a combination of both. Network has great importance in gateway load balancing mechanism. The gateway node plays an important role as a bridge between different networks domains. The traffic load in gateways is a crucial part of the network which needs to be resolved and are important in order to reduce the load imbalance and to maximize the total network throughput. Such gateway load balancing is considered to improve the network performance more effectively than multipath load balancing. In this seminar, I am trying to show the network traffic, importance of traffic management in network especially in the gateways and point out that performing gateway load balancing is vital in order to take full advantage of the resources available and thereby improve the performance. A research study in this area and methodology to identify gateway load-balancing and characterize load-balanced paths is performed.

Ant Based Survivable Routing and Load Balancing Technique in WDM Networks

In Wavelength Division Multiplexing (WDM) networks, providing survivability during routing along with maintaining load balancing is a daunting task. In this paper, we propose an Ant based survivable routing and load balancing technique in WDM networks which discovers survivable and load balanced feasible paths. Ant agents are launched in the network to gather essential information such as available bandwidth, route length and number of free wavelengths. A combined feasibility value is generated from these factors for each path. While transmitting data to the destination, the source selects the path with high feasible value and low load as primary path, and other paths as backup paths. During congestion and failures, data transmission is redirected through the backup paths. Thus, our technique provides survivability and load balancing at hand efficiently. Simulation results show that our technique reduces packet drops efficiently by detecting and controlling the congestion.

Revisiting flow-based load balancing: Stateless path selection in data center networks

Hash-based load-balancing techniques are widely used to distribute the load over multiple forwarding paths and preserve the packet sequence of transport-level flows. Forcing a long-lived, i.e., elephant, flow to follow a specific path in the network is a desired mechanism in data center networks to avoid crossing hot spots. This limits the formation of bottlenecks and so improves the network use. Unfortunately, current per-flow load-balancing methods do not allow sources to deterministically force a specific path for a flow.In this paper, we propose a deterministic approach enabling end hosts to steer their flows over any desired load-balanced path without relying on any packet header extension. By using an invertible mechanism instead of solely relying on a hash function in routers, our method allows to easily select the packet’s header field values in order to force the selection of a given load-balanced path without storing any state in routers.We perform various simulations and experiments to evaluate the performance and prove the feasibility of our method using a Linux kernel implementation. Furthermore, we demonstrate with simulations and lab experiments how MultiPath TCP can benefit from the combination of our solution with a flow scheduling system that efficiently distributes elephant flows in large data center networks.