Load Balancing Scheme Research Articles

RD-FCA: A resilient distributed framework for formal concept analysis

Formal Concept Analysis (FCA) is a data analysis technique with applications in data mining, artificial intelligence, software engineering, etc. The algorithms for FCA are computationally expensive, and their recursion tree is highly irregular and dynamic in nature. Several distributed FCA algorithms have been proposed to exploit parallelism within and across machines. However, none of the distributed approaches are able to recover from failures in the system. We propose RD-FCA, the first resilient distributed framework for FCA that uses novel load-balancing strategy, handles fail-stop failures and provides at-least-once semantics for concept discovery. Our asynchronous snapshot mechanism with incremental updates reduces the snapshot overhead and minimizes recalculation of concepts during recovery. RD-FCA also supports dynamic addition of workers to accelerate performance. Compared to MapReduce based approaches, RD-FCA performs an order of magnitude faster. We show through extensive evaluation that RD-FCA recovers efficiently from single, multiple, independent and cascading failures.

Distributed Differential Evolution With Adaptive Resource Allocation.

Distributed differential evolution (DDE) is an efficient paradigm that adopts multiple populations for cooperatively solving complex optimization problems. However, how to allocate fitness evaluation (FE) budget resources among the distributed multiple populations can greatly influence the optimization ability of DDE. Therefore, this article proposes a novel three-layer DDE framework with adaptive resource allocation (DDE-ARA), including the algorithm layer for evolving various differential evolution (DE) populations, the dispatch layer for dispatching the individuals in the DE populations to different distributed machines, and the machine layer for accommodating distributed computers. In the DDE-ARA framework, three novel methods are further proposed. First, a general performance indicator (GPI) method is proposed to measure the performance of different DEs. Second, based on the GPI, a FE allocation (FEA) method is proposed to adaptively allocate the FE budget resources from poorly performing DEs to well-performing DEs for better search efficiency. This way, the GPI and FEA methods achieve the ARA in the algorithm layer. Third, a load balance strategy is proposed in the dispatch layer to balance the FE burden of different computers in the machine layer for improving load balance and algorithm speedup. Moreover, theoretical analyses are provided to show why the proposed DDE-ARA framework can be effective and to discuss the lower bound of its optimization error. Extensive experiments are conducted on all the 30 functions of CEC 2014 competitions at 10, 30, 50, and 100 dimensions, and some state-of-the-art DDE algorithms are adopted for comparisons. The results show the great effectiveness and efficiency of the proposed framework and the three novel methods.

A novel two-phase energy efficient load balancing scheme for efficient data collection for energy harvesting WSNs using mobile sink

A novel two-phase energy efficient load balancing scheme for efficient data collection for energy harvesting WSNs using mobile sink

Artificial Intelligence Empowered Traffic Control for Internet of Things with Mobile Edge Computing

Mobile edge computing (MEC) is one of the efficient technologies to provide satisfying quality of experience (QoE) for emerging computation-intensive applications in internet of things (IoT). However, some new challenges will be encountered when MEC is applied in a large-scale IoT with massive access devices or heavy traffic loads such as load balancing and traffic offloading. Aiming at the solution of these problems, this paper proposes a learning-based traffic control architecture for IoT with MEC. Moreover, a deep-learning-based load balancing framework is developed to control user association in IoT. The user association is determined at each IoT access points (IAP) by the deep neural network (DNN), which is the duplication of the well-trained DNN with the global network information. In addition, we propose a reinforcement-learning-based partial traffic offloading scheme to reduce the traffic origination. The IoT devices are able to independently decide its offloading radio according to the channel quality information, service requirement, and workload of the IAP. Simulation results indicate that the proposed deep-learning-based load balancing scheme is able to achieve uniform traffic distribution, and meanwhile our partial offloading policy can significantly reduce the network traffic.

ESCALB: An effective slave controller allocation-based load balancing scheme for multi-domain SDN-enabled-IoT networks

In software-defined networking (SDN), several controllers improve the reliability as well as the scalability of networks such as the Internet-of-Things (IoT), with the distributed control plan. To achieve optimal results in IoT networks, an SDN can be employed to reduce the complexity associated with IoT and provide an improved quality-of-service (QoS). With time, it is likely expected that the demand for IoT will rise, and a large number of sensors will be connected, which can generate huge network traffic. With these possibilities, the SDN controllers processing capacity will be surpassed by the traffic sent by the IoT sensors. To handle this kind of challenge, and achieve promising results, a dynamic slave controller allocation with a premeditated mechanism can play a pivotal role to accomplish the task management and migration plan. Following this, we proposed an efficient slave controller allocation-based load balancing approach for a multi-domain SDN-enabled IoT network, which aims to transfer switches to a controller with idle resources effectively. Among several slave controllers for selecting a target controller, a multi-criteria decision-making (MCDM) strategy, i.e., an analytical network process (ANP) has been used in our approach to enrich communication metrics and maintain high-standard QoS statistics. Moreover, switch migration is modeled with knapsack 0/1 problem to achieve maximum utilization of the slave controllers. Our proposed scheme enabled with a flexible decision-making process for selecting controllers with varying resources. The results demonstrated with emulation environment show the effectiveness of the ESCALB.

Application study of ant colony algorithm for network data transmission path scheduling optimization

Abstract With the rapid development of the information age, the traditional data center network management can no longer meet the rapid expansion of network data traffic needs. Therefore, the research uses the biological ant colony foraging behavior to find the optimal path of network traffic scheduling, and introduces pheromone and heuristic functions to improve the convergence and stability of the algorithm. In order to find the light load path more accurately, the strategy redefines the heuristic function according to the number of large streams on the link and the real-time load. At the same time, in order to reduce the delay, the strategy defines the optimal path determination rule according to the path delay and real-time load. The experiments show that under the link load balancing strategy based on ant colony algorithm, the link utilization ratio is 4.6% higher than that of ECMP, while the traffic delay is reduced, and the delay deviation fluctuates within ±2 ms. The proposed network data transmission scheduling strategy can better solve the problems in traffic scheduling, and effectively improve network throughput and traffic transmission quality.

An energy-efficient load balance strategy based on virtual machine consolidation in cloud environment

In the cloud computing environment, unbalanced utilization of multi-dimensional resources in physical servers generates resource fragmentation, leading to inefficient resource utilization and energy wastage in data centers. The inefficient utilization of resources makes high energy consumption and low quality of service (QoS) in resource management an urgent problem to be solved. We propose a load balancing strategy based on virtual machine consolidation (LBVMC), which aims to reduce the energy consumption and service level agreement (SLA) violation of data centers by balancing the multi-dimensional resource utilization in physical machines (PMs). First, we present a load state classification algorithm for PMs with load abnormality considering current and future loads to reduce unnecessary virtual machine (VM) migrations caused by occasional load fluctuations. Then, we propose a resource-weight based selection model for migratable VMs, which selects appropriate VMs to be migrated based on multi-dimensional resource utilization, and reduces resource fragmentation caused by load imbalance. Finally, we design a VM placement algorithm based on resource fitness and load correlation to deploy VMs on the optimal destination PMs to assure load balancing of the destination PMs after VM placement. We perform simulated experiments under homogeneous, heterogeneous, and bottleneck resource environments. Experimental results show that LBVMC is superior to other strategies in reducing energy consumption and SLA violations and achieves better overall performance.

An Intelligent Server load balancing based on Multi-criteria decision-making in SDN

In an environment of rising internet usage, it is difficult to manage network traffic while maintaining a high quality of service. In highly trafficked networks, load balancers are crucial for ensuring the quality of service. Although different approaches to load-balancing have been proposed in traditional networks, some of them require manual reconfiguration of the device to accommodate new services due to a lack of programmability. These problems can be solved through the use of software-defined networks. This research paper presents a dynamic load-balancing algorithm for software-defined networks based on server response time and content mapping. The proposed technique dispatches requests to servers based on real-time server loads. This technique comprises three different modules, such as a request classification module, a server monitoring module, and an optimized dynamic load-balancing module using content-based routing. There are a variety of robust mathematical tools to address complex problems that have multiple objectives. Multi-Criteria Decision-Making is one of them. The performance of the proposed scheme has been validated by applying the Weighted Sum Method of the multi-criteria decision-making technique. The proposed method Server load balancing based on Multi-criteria Decision Making[SDLB-MCDM] is compared with different load-balancing schemes such as round robin, random, load-balancing scheme based on server response time [LBBSRT], and An SDN-aided mechanism for web load- balancing based on server statistics [SD-WLB]. The experimental results of SDLB-MCDM show a significant improvement of 58% when weights are equal and 50% when unequal weights are assigned to various QoS parameters in comparison with the ROUND ROBIN, RANDOM, LBBSRT and SD-WLB techniques.

Multi-level parallelization of quantum-chemical calculations.

Strategies for multiple-level parallelizations of quantum-mechanical calculations are discussed, with an emphasis on using groups of workers for performing parallel tasks. These parallel programming models can be used for a variety abinitio quantum chemistry approaches, including the fragment molecular orbital method and replica-exchange molecular dynamics. Strategies for efficient load balancing on problems of increasing granularity are introduced and discussed. A four-level parallelization is developed based on a multi-level hierarchical grouping, and a high parallel efficiency is achieved on the Theta supercomputer using 131072 OpenMP threads.

An Efficient Switch Migration Scheme for Load Balancing in Software Defined Networking

Software-defined networking (SDN) provides increased flexibility to network management through distributed SDN control, and it has been a great breakthrough in network innovation. Switch migration is extensively used for workload balancing among distributed controllers. The time-sharing switch migration (TSSM) scheme proposes a strategy in which more than one controller is allowed to share the workload of a switch via time sharing during overloaded conditions, resulting in the mitigation of ping-pong controller difficulty, a reduced number of overload occurrences, and better controller efficiency. However, it has increased migration costs and higher controller resource consumption during the TSSM operation period because it requires more than one controller to perform. Therefore, we have proposed a strategy that optimizes the controller selection during the TSSM period based on flow characteristics through a greedy set coverage algorithm. The improved TSSM scheme provides reduced migration costs and lower controller resource consumption, as well as TSSM benefits. For its feasibility, the implementation of the proposed scheme is accomplished through an open network operating system. The experimental results show that the proposed improved TSSM scheme reduces the migration cost and lowers the controller resource consumption by about 36% and 34%, respectively, as compared with the conventional TSSM scheme.

Automatically balancing relocatable distributed collections

SummaryIn previous work, we introduced a distributed collections library for the APGAS for Java programming model. This library makes it possible for programmers to develop complex distributed programs thanks to the many abstractions and computation patterns supported. In particular, programmers can fully and dynamically change the distribution of data entries through high‐level abstractions. However, the problem of balancing the load between processes remains, especially in cases where multiple processes may be concurrently executing on a single host, or when the performance of the hosts used differs. To address this issue and to relieve the burden of programming a load balancing strategy for a specific application, we created a dynamic load balancer integrated into our library. This load balancer operates within a specific context in a manner which does not interfere with the program legibility. Internally, we implement a scheme inspired by the lifeline‐based global load balancer scheme first introduced in 10. We evaluate the performance of our integrated load balancer on a small‐scale Beowulf cluster.

A B5G Non-Terrestrial-Network (NTN) and Hybird Constellation Based Data Collection System (DCS)

In beyond 5G (B5G) non-terrestrial network (NTN) systems, satellite technologies play an important role. Especially for data collection systems (DCS), low-earth orbit satellites have many advantages. Such as global coverage, low latency, and high efficiency. As a miniaturization technology, CubeSat has attracted extensive attention from a large number of scholars. Satellite constellations can coordinate for distributed tasks. This paper proposes a B5G NTN-based data collection system. A CubeSat constellation achieves global coverage as the basic space platform for DCS. The 5G terrestrial network is used as the data bearer network of the gateway station. A traffic load balance strategy is proposed to optimize the system’s efficiency. As a unified hardware platform, software-defined radio (SDR) is compatible with various sensor data models. Finally, the design was verified by a series of experiments.

Efficient Load Balancing in WSN Using Quasi – oppositional Based Jaya Optimization with Cluster Head Selection

Researchers have been paying close attention to the wireless sensor (WSN) networks area because of its variety of applications, including industrial management, human detection, and health care management. In Wireless Sensor (WSN) Network, consumption of efficient energy is a challenging problem. Many clustering techniques were used for balancing the load of WSN network. In clustering, the cluster head (CH) is selected as a relay node with greater power which is compared with the nodes of non-CH. In the existing system, it uses LBC-COFL algorithm to reduce the energy consumption problem. To overcome this problem, the proposed system uses Quasi oppositional based Jaya load balancing strategy with cluster head (QOJ-LCH) selection protocol to boost the lifespan of network and energy consumption. The QOJ-LCH method improves the relay nodes life and shares the load on relay nodes equitably across the network to enhance the lifespan. It also reduces the load-balancing problems in Wireless Sensor networks. It uses two routing methods single-hop and multiple-hop. The proposed QOJ-LCH with cluster head selection method enhances the network’s lifespan, total amount of power utilization and the active sensor devices present in the Single-hop routing,it worked with 1600 rounds in network and 300 sensor nodes, for Multiple-hop routing, it worked with 1800 rounds in network and 350 sensor nodes. It achieves better performance, scalability and reliability.

Lifeline-based load balancing schemes for Asynchronous Many-Task runtimes in clusters

A popular approach to program scalable irregular applications is Asynchronous Many-Task (AMT) Programming. Here, programs define tasks according to task models such as dynamic independent tasks (DIT) or nested fork-join (NFJ). We consider cluster AMTs, in which a runtime system maps the tasks to worker threads in multiple processes.Thereby, dynamic load balancing can be achieved via cooperative work stealing, coordinated work stealing, or work sharing. A well-performing cooperative work stealing variant is the lifeline scheme. While previous implementations of this scheme are restricted to single-worker processes, a recent hybrid extension combines it with intra-process work sharing between multiple workers. The hybrid scheme, which was proposed for both DIT and NFJ, comes at the price of a higher complexity.This paper investigates whether this complexity is indispensable for multi-worker processes by contrasting the hybrid scheme with a novel pure work stealing extension of the lifeline scheme to multiple workers. We independently implemented the extension for DIT and NFJ. In experiments based on four benchmarks, we observed the pure scheme to be on a par or even outperform the hybrid one by up to 18% for DIT and up to 5% for NFJ.Building on this main result, we studied a modification of the pure scheme, which prefers local over global victims, and more heavily loaded over less loaded ones. The modification improves the performance of the pure scheme by up to 15%. Finally, we explored whether the lifeline scheme can profit from a change to coordinated work stealing. We developed a coordinated multi-worker implementation for DIT and observed a performance improvement over the cooperative scheme by up to 17%.

Hybrid grey wolf and improved particle swarm optimization with adaptive intertial weight-based multi-dimensional learning strategy for load balancing in cloud environments

The popularization of Internet applications and rapid advent of information technologies have invited increased number of developers and companies to focus on the area of Cloud Computing (CC). The most significant issues and challenges in the domain of CC include Load Balancing (LB), scheduling of task executions and managing resource allocation. In particular, LB being the process of distributing computing resources and workloads has attracted maximum attention as it is the predominant issue in CC. In this paper, Hybrid Grey Wolf and Improved Particle Swarm Optimization Algorithm with Adaptive Intertial Weight-based multi-dimensional Learning Strategy (HGWIPSOA)-based LB scheme is proposed for enhancing precision and rapidness in task scheduling and assignment of resources to Virtual Machines (VMs) in cloud environments. In the proposed scheme, initially, Grey Wolf Optimization Algorithm (GWOA) is incorporated into Particle Swarm Optimisation (PSO) for considering the highest fitness particle as alpha wolf search agent, such that the objective of allocating tasks to VMs is attained effectively and efficiently. It then integrates chaos, Adaptive Inertial Weight (AIW) and Dimensional Learning (DL) into PSO specifically to prevent premature convergence and achieve better convergence speed and global search ability depending on the best experience determined by particles for effective LB. The simulation experiments of the proposed HGWIPSOA mechanism confirm better results by offering 21.32 % improved throughput, 19.84 % reduced makespan, 24.98 % minimized degree of imbalance, 18.74 % reduced latency and 27.92 % reduced execution time independent of the number of tasks in the cloud environment on par with benchmarked LB schemes.

New Improved Multi-Objective Gorilla Troops Algorithm for Dependent Tasks Offloading problem in Multi-Access Edge Computing

Computational offloading allows lightweight battery-operated devices such as IoT gadgets and mobile equipment to send computation tasks to nearby edge servers to be completed, which is a challenging problem in the multi-access edge computing (MEC) environment. Numerous conflicting objectives exist in this problem; for example, the execution time, energy consumption, and computation cost should all be optimized simultaneously. Furthermore, offloading an application that consists of dependent tasks is another important issue that cannot be neglected while addressing this problem. Recent methods are single objective, computationally expensive, or ignore task dependency. As a result, we propose an improved Gorilla Troops Algorithm (IGTA) to offload dependent tasks in the MEC environments with three objectives: 1-Minimizing the execution latency of the application, 2-energy consumption of the light devices, 3-the used cost of the MEC resources. Furthermore, it is supposed that each MEC supports many charge levels to provide more flexibility to the system. Additionally, we have extended the operation of the standard Gorilla Troops Algorithm (GTO) by adopting a customized crossover operation to improve its search strategy. A Max-To-Min (MTM) load-balancing strategy was also implemented in IGTA to improve the offloading operation. Relative to GTO, IGTA has reduced latency by 33%, energy consumption by 93%, and cost usage by 34.5%. We compared IGTA with other Optimizers in this problem, and the results showed the superiority of IGTA.

Load Balancing Strategy of Power Communication Network Based on SDN Controller

A load balancing strategy based on SDN controller is proposed to solve the load and security problems caused by the large number of services carried by power communication network in this article. Firstly, integrated with the actual situation of the network, the basic framework of gradual hierarchical transformation of power communication network is put forward. Then, the SDN controller response delay model is discussed in detail in the modeling process, and a multi controller deployment model and its solution algorithm are also proposed. The simulation results verify the effectiveness of the SDN centralized control function, and optimize the load balance, QoS and resource utilization, so as to reduce the impact of power grid communication link failure on the control delay.

Load balancing inside programmable data planes based on network modeling prediction using a GNN with network behaviors

In data center networks, existing control plane- and end host-based load-balancing methods are encumbered by excessively large decision delays during rapid reactions to microbursts. However, existing programmable data plane-based load balancing methods require large overheads involved in active probing. Accurate network modeling can optimize load balancing. However, existing modeling methods suffer from low generalization and high overhead. In this study, we propose a network modeling method based on a graph neural network (GNN) with basic network behavior (hereinafter called GNN-Behavior). This is derived from two inherent correlations observed: the correlation between global network behavior and basic network behavior and the correlation among basic network behaviors. We employed the GNN with an improved message-passing neural network to learn such two inherent correlations. Particularly, we considered modeling end-to-end delay as a use case to validate GNN-Behavior. Furthermore, we propose a packet-level load-balancing scheme inside programmable data planes (PDPs) based on the accurate prediction of end-to-end delay from the GNN-Behavior model(LBPP). LBPP is a control plane-PDP collaborative method that integrates a global view from a controller and quick response from switches. Experimental results demonstrate the feasibility and effectiveness of the GNN-Behavior and LBPP. Compared with queuing theory (QT), RouteNet, and GNN-based scheme, GNN-Behavior increases goodness of fit (R2) by 73.1%, 11.1%, and 3.74%, respectively. Under an unknown traffic control strategy, the generalization ability of GNN-Behavior is considerably better than that of QT and RouteNet. Compared with flow-level ECMP, flowlet-level LetFlow, and packet-level DRILL, LBPP can reduce average flow completion time by up to 43.9%, 37.4%, and 17.2%, respectively.

Load balancing in 5G heterogeneous networks based on automatic weight function

Load balancing is a major challenge in heterogeneous networks (HetNets) consisting of 5G and 6G ultra-dense small cells with long-term evaluation advanced (LTE-A) networks. A key factor in achieving efficient load balancing during user mobility is creating appropriate optimisation for handover control parameters (HCP). This paper proposes a coordinated load balancing algorithm for LTE-A/fifth generation (5G) HetNets. The algorithm automatically optimises HCP settings for a given user based on three bounded functions (the signal-to-interference-plus-noise ratio (SINR) of the user equipment (UE), the number of physical resource blocks (PRBs) per UE and the UE’s speed) as well as their automatic weight levels. A two-step target cell determination strategy is implemented according to the cell load level and RSRP criteria, ensuring that users are handed over to low-loaded target cells. A new HO procedure that considers the pilot signal power is also proposed, which includes the number of PRBs per UE and the RSRP. Cells with freer PRBs are prioritised in user association to provide load balance and enhanced throughput. The proposed load balancing algorithm is compared with five other load balancing algorithms selected from the literature. The simulation results reveal that under various mobile speed scenarios, the proposed load balancing scheme enhances network performance in terms of load level, throughput, spectral efficiency and call dropping ratio (CDR).

Hybrid dingo and whale optimization algorithm‐based optimal load balancing for cloud computing environment

AbstractCloud computing is the most rapidly growing technology that facilitates virtualized computer resources to the users through its service providers. Load balancing and task scheduling is the important concern in the cloud computing scenario from the service provider point of view. The process of load balancing is responsible for allocating the dynamic workload among the cloud systems, such that equal sharing of the resources without the virtual machines (VMs) being overloaded or underloaded. This load balancing process not only needs to allocate the tasks to suitable VMs but also need to maintain trade‐off among VMs. Thus, a potential dynamic and optimized load balancing strategy is essential for improving resource utilization, services, reliability, and throughput by maintaining equilibrium between VMs with minimized makespan. In this article, hybrid dingo and whale optimization algorithm‐based load balancing mechanism (HDWOA‐LBM) is proposed for effective load balancing those aids in maximized throughput, reliability, and resource utilization in the clouds. This HDWOA‐LBM is proposed by mimicking the hunting characteristics of dingo (equivalent to tasks) and VMs as their prey to be hunted. It inherited the process of exploration and exploitation through the chasing and approaching, encircling, and harassing inherent with dingo optimization algorithm (DOA) to determine the optimal allocation of incoming tasks to the suitable VM. In specific, it utilized the merits of whale optimization for improving the exploitation phase of DOA to balance the trade‐off between local and global search. The simulation experiments of the proposed HDWOA‐LBM conducted using CloudSim confirmed better throughput of 21.28%, maximized reliability of 25.42%, minimized makespan of 22.98%, and improved resource allocation of 20.86%, on par with the competitive intelligent load balancing schemes used for investigation.