Dynamic Load Balancing Technique Research Articles

Dynamic Q-Learning-Based Optimized Load Balancing Technique in Cloud

Cloud computing provides on-demand access to a shared puddle of computing resources, containing applications, storage, services, and servers above the internet. This allows organizations to scale their IT infrastructure up or down as needed, reduce costs, and improve efficiency and flexibility. Improving professional guidelines for social media interactions is crucial to address the wide range of complex issues that arise in today’s digital age. It is imperative to enhance and update professional guidelines regarding social media interactions in order to effectively tackle the multitude of intricate issues that emerge. In this paper, we propose a reinforcement learning (RL) method for handling dynamic resource allocation (DRA) and load balancing (LB) activity in a cloud environment and achieve good scalability and a significant improvement in performance. To address this matter, we propose a dynamic load balancing technique based on Q-learning, a reinforcement learning algorithm. Our technique leverages Q-learning to acquire an optimal policy for resource allocation in real-time based on existing workload, resource accessibility, and user preferences. We introduce a reward function that takes into account performance metrics such as response time and resource consumption, as well as cost considerations. We evaluate our technique through simulations and show that it outperforms traditional load balancing techniques in expressions of response time and resource utilization while also reducing overall costs. The proposed model has been compared with previous work, and the consequences show the significance of the proposed work. Our model secures a 20% improvement in scalability services. The DCL algorithm offers significant advantages over genetic and min-max algorithms in terms of training time and effectiveness. Through simulations and analysis on various datasets from the machine learning dataset repository, it has been observed that the proposed DCL algorithm outperforms both genetic and min-max algorithms. The training time can be reduced by 10% to 45%, while effectiveness is enhanced by 30% to 55%. These improvements make the DCL algorithm a promising option for enhancing training time and effectiveness in machine learning applications. Further research can be conducted to investigate the potential of combining the DCL algorithm with a supervised training algorithm, which could potentially further improve its performance and apply in real-world application.

Targeting a faster time‐to‐solution of mortar‐based contact problems

AbstractWhile mortar methods are undoubtedly among the most prominent and accurate approaches in computational contact mechanics, they come with considerable cost. The required numerical effort originates not only from the evaluation of the mortar integrals but also from the solution of the system of linear equations within Newton‐type solvers. While each of these cost factors has been analyzed and addressed individually in the past, this contribution outlines approaches and first results to comprehensively reduce the time to solution by combining dynamic load balancing techniques with algebraic multigrid preconditioners. We present a proof of concept for a three‐dimensional unilateral contact simulation with an evolving contact zone.

MLPRS: A Machine Learning-Based Proactive Re-Routing Scheme for flow classification and priority assignment

In modern networking scenarios, the transmission of numerous burst packets is a common requirement. In this context, Software Defined Networks (SDNs) have been demonstrated to be an effective solution with a programmable centralized controller for managing complex networks. This article aims to propose a Machine Learning-based Proactive Re-routing Scheme (MLPRS) that aims to enhance the Quality of Service (QoS) by dynamic load balancing in real-time network topology. The proposed scheme applies machine learning (ML) techniques to classify applications and assign flow priorities based on their type. The betweenness centrality algorithm is used to identify the importance of each link in the network, and the links are ordered in the betweenness set. The proposed method monitors the load on each link and if the critically important links are overloaded, then the flow is routed in real-time to avoid congestion. The idea of MLPRS has been simulated in Mininet with an OpenDaylight controller, and the results show that it effectively improves QoS and network performance. The MLPRS approach contributes to the ongoing efforts in developing intelligent and dynamic load-balancing techniques using machine learning algorithms. The performance of MLPRS has been improved by 3.7 % and 3.6 % in terms of Throughput and Bandwidth respectively when compared with the default SDN.

Improved round-robin rule learning for optimal load balancing in distributed cloud systems

A newly emerging technology that distributes virtualised computer resources across the internet is referred to as 'cloud computing' and is gaining popularity. These clouds' ability to evenly distribute the load is vital. Load balancing in cloud computing distributes dynamic workloads so no database server is overcrowded or underloaded (LB). As a result, a dynamic load balancing technique in the cloud may contribute to improved service stability and resource utilisation. In this study, we model a load balancing system that balances the cloud's available resources using a rule-based round-robin method. Cloud parameters are used to calculate and allocate resources using a rule-based round-robin technique. The study analyses cloud metrics to determine cloud loads and resources. The cloud sim tool is used to allocate resources, and its effectiveness is tested against a variety of performance measures such as overhead, migration time, and throughput rate. Simulation results show that rule-based round-robin improves cloud performance. When compared to existing algorithms, the proposed methodology improves task performance by about 4%.

Dynamic Load Balancing Techniques in the IoT: A Review

The Internet of things (IoT) extends the Internet space by allowing smart things to sense and/or interact with the physical environment and communicate with other physical objects (or things) around us. In IoT, sensors, actuators, smart devices, cameras, protocols, and cloud services are used to support many intelligent applications such as environmental monitoring, traffic monitoring, remote monitoring of patients, security surveillance, and smart home automation. To optimize the usage of an IoT network, certain challenges must be addressed such as energy constraints, scalability, reliability, heterogeneity, security, privacy, routing, quality of service (QoS), and congestion. To avoid congestion in IoT, efficient load balancing (LB) is needed for distributing traffic loads among different routes. To this end, this survey presents the IoT architectures and the networking paradigms (i.e., edge–fog–cloud paradigms) adopted in these architectures. Then, it analyzes and compares previous related surveys on LB in the IoT. It reviews and classifies dynamic LB techniques in the IoT for cloud and edge/fog networks. Lastly, it presents some lessons learned and open research issues.

A Review on Load Balancing Algorithms in Cloud Computing

Cloud computing is a new standard for large-scale distributed and parallel computing that is still in its infancy. At a predetermined time, provides common resources, information, software packages, and other resources in accordance with client requirements. As cloud computing becomes more popular and more people become interested in assistive computing, better and quicker service is necessary. It is necessary to enhance the management of the good load balancing solutions that are already accessible. Recently, numerous clients from all over the world have been seeking various services at a quick rate, negatively impacting cloud computing performance. The various load balancing techniques used in cloud computing are effective in customizing demand by identifying the appropriate virtual machines. In this review, a comparative study will be done between many load balancing algorithms in different cloud computing, as many studies have been done on this important topic. Important research will also be analyzed and defined, and analyzes will be written for each research that talks about dynamic and non-dynamic load balancing techniques and algorithms. Many details will also be mentioned from the researcher's point of view as a summary of each research.

In-depth analysis of dynamic degree load balancing technique in public cloud for heterogeneous cloudlets

<span lang="EN-US">Load <span lang="EN-US">balancing is one of the challenges of the distributed computing worldview. With the enormous development in clients and their interest for different administrations on the distributed computing stage, compelling or productive asset usage in the cloud climate has turned into an urgent concern. Load balancing is critical to keeping cloud computing running smoothly. This study examines the research using four scheduling algorithms: dynamic degree balance CPU based (D2B_CPU), dynamic degree balanced membership based (D2B_Membership), dynamic degree memory balanced allocation (D2MBA) and hybrid dynamic degree balance (HDDB) algorithm. Central processing unit (CPU) utilisation, bandwidth utilisation, and memory utilisation are used as performance measures to verify the performance of these algorithms. The CloudSim simulation programme was used to simulate these algorithms. The primary goal of this work is to aid in the future construction of new algorithms by researching the behaviour of various existing algorithms.</span></span>

Empirical analysis of dynamic load balancing techniques in cloud computing.

Virtualization, dispersed registration, systems administration, programming, and web administrations are all examples of "distributed computing." Customers, datacenters, and scattered servers are just a few of the components that make up a cloud. It includes things like internal failure adaption, high accessibility, flexibility, adaptability, lower client overhead, lower ownership costs, on-demand advantages, and so on. The basis of a feasible load adjusting computation is key to resolving these challenges. CPU load, memory limit, deferral, and system load are all examples of heaps. Burden adjustment is a method for distributing the load across the many hubs of a conveyance framework in order to optimize asset utilization and employment response time while avoiding a situation where some hubs are heavily loaded while others are idle or performing little work. Burden adjustment ensures that at any one time, each processor in the framework or each hub in the system does about the same amount of work. This method may be initiated by the sender, the collector, or the symmetric sort (the blend of sender-started and recipient started types). With some example data center loads, the goal is to create several dynamic load balancing techniques such as Round Robin, Throttled, Equally Spread Current Execution Load, and Shortest Job First algorithms.

Cloud Dynamic Load Balancing and Reactive Fault Tolerance Techniques: A Systematic Literature Review (SLR)

The computational cloud aims to move traditional computing from personal computers to cloud providers on the internet. Cloud security represents an important research area. Confidentiality, integrity, and availability are the main cloud security characteristics addressed. Cloud providers apply dynamic load balancing and reactive fault tolerance techniques to build secure cloud services to achieve high service availability. Dynamic cloud load balancing approaches distribute submitted tasks to virtual machines during tasks execution and the load of virtual machines is updated based on the system’s state. Reactive cloud fault tolerance is activated for system process failures after failure effectively happens. Reactive cloud fault tolerance handles failure after the fault has occurred. Despite the significance of dynamic load balancing and reactive fault tolerance techniques and mechanisms, few reviews focus on these approaches in a systematic, unbiased method focusing on integrating cloud dynamic load balancing and reactive fault tolerance techniques. This paper conducts a systematic literature review of the existing literature concerning reactive fault tolerance, dynamic load balancing, and their integration in their basic approaches, types, frameworks, and future directions.

Adaptive Neuro Fuzzy Interference and PNN Memory Based Grey Wolf Optimization Algorithm for Optimal Load Balancing

In recent years, cloud computing provides a spectacular platform for numerous users with persistent and alternative varying requirements. In the cloud environment, security and service availability are the two most significant factors during the data encryption process. For providing optimal service availability, it is necessary to establish a load balancing technique that is capable of balancing the request from diverse nodes present in the cloud. This paper aims in establishing a dynamic load balancing technique using the APMG approach. Here in this paper, we integrated adaptive neuro-fuzzy interference system-polynomial neural network as well as memory-based grey wolf optimization algorithm for optimal load balancing. The memory-based grey wolf optimization algorithm is employed to enhance the precision of ANFIS-PNN and to maximize the locations of the membership functions respectively. Also, two significant factors namely the turnaround time and CPU utilization involved in optimal load balancing scheme are evaluated. Finally, the performance evaluation of the proposed MG-ANFIS based dynamic load balancing approach is compared with various other load balancing approaches to determine the system performances.

CMODLB: an efficient load balancing approach in cloud computing environment

A hybrid of supervised (artificial neural network), unsupervised (clustering) machine learning, and soft computing (interval type 2 fuzzy logic system)-based load balancing algorithm, i.e., clustering-based multiple objective dynamic load balancing technique (CMODLB), is introduced to balance the cloud load in the present work. Initially, our previously introduced artificial neural network-based dynamic load balancing (ANN-LB) technique is implemented to cluster the virtual machines (VMs) into underloaded and overloaded VMs using Bayesian optimization-based enhanced K-means (BOEK-means) algorithm. In the second stage, the user tasks are scheduled for underloading VMs to improve load balance and resource utilization. Scheduling of tasks is supported by multi-objective-based technique of order preference by similarity to ideal solution with particle swarm optimization (TOPSIS-PSO) algorithm using different cloud criteria. To realize load balancing among PMs, the VM manager makes decisions for VM migration. VM migration decision is done based on the suitable conditions, if a PM is overloaded, and if another PM is minimum loaded. The former condition balances load, while the latter condition minimizes energy consumption in PMs. VM migration is achieved through interval type 2 fuzzy logic system (IT2FS) whose decisions are based on multiple significant parameters. Experimental results show that the CMODLB method takes 31.067% and 71.6% less completion time than TaPRA and BSO, respectively. It has maintained 65.54% and 68.26% less MakeSpan than MaxMin and R.R algorithms, respectively. The proposed method has achieved around 75% of resource utilization, which is highest compared to DHCI and CESCC. The use of novel and innovative hybridization of machine learning, multi-objective, and soft computing methods in the proposed algorithm offers optimum scheduling and migration processes to balance PMs and VMs.

A Dynamic Multi–Objective Approach for Dynamic Load Balancing in Heterogeneous Systems

Modern standards in High Performance Computing (HPC) have started to consider energy consumption and power draw as a limiting factor. New and more complex architectures have been introduced in HPC systems to afford these new restrictions, and include coprocessors such as GPGPUs for intensive computational tasks. As systems increase in heterogeneity, workload distribution becomes a more core problem to achieve the maximum efficiency in every computational component. We present a Multi-Objective Dynamic Load Balancing (DLB) approach where several objectives can be applied to tune an application. These objectives can be dynamically exchanged during the execution of an algorithm to better adapt to the resources available in a system. We have implemented the Multi-Objective DLB together with a generic heuristic engine, designed to perform multiple strategies for DLB in iterative problems. We also present Ull Multiobjective Framework (UllMF), an open-source tool that implements the Multi-Objective generic approach. UllMF separates metric gathering, objective functions to be optimized and load balancing algorithms, and improves code portability using a simple interface to reduce the costs of new implementations. We illustrate how performance and energy consumption are improved for the implemented techniques, and analyze their quality using different DLB techniques from the literature.

A dynamically load-balanced parallel p-adaptive implicit high-order flux reconstruction method for under-resolved turbulence simulation

We present a dynamically load-balanced parallel p-adaptive implicit high-order flux reconstruction method for under-resolved turbulence simulation. The high-order explicit first stage, singly diagonal implicit Runge–Kutta (ESDIRK) method is employed to circumvent the restriction on the time step size. The pseudo transient continuation is coupled with the matrix-free restarted generalized minimal residual (GMRES) method to solve the nonlinear equations at each stage, except the first one, of ESDIRK. We use the spectral decay smoothness indicator as the refinement/coarsening indicator for p-adaptation. A dynamic load balancing technique is developed with the aid of the open-source library ParMETIS. The trivial cost, compared to implicit time stepping, of mesh repartitioning and data redistribution enables us to conduct p-adaptation and load balancing every time step. An isentropic vortex propagation case is employed to study the impact of element weights used in mesh repartitioning on parallel efficiency. We apply the p-adaptive solver for implicit large eddy simulation (ILES) of the transitional flows over a cylinder when Reynolds number (Re) is 3900 and the SD7003 wing when Re is 60000. Numerical experiments demonstrate that a significant reduction in the run time (up to 70%) and total number of solution points (up to 76%) can be achieved with p-adaptation.

Load Balancing Problem on Hyper Hexa Cell Interconnection Network

Dynamic load balancing techniques prevents computer nodes from overloading unevenly while leaving other idle. It is considered as one of the most challenging topics in parallel computing. Moreover, it is essential for increasing the efficiency of highly parallel systems especially in solving multitask problems with unpredictable load estimates. Particularly, over each processor in the parallel systems and interconnection networks. This paper focuses on developing an efficient algorithm for load balancing on Hyper Hexa Cell (HHC) interconnection network, namely, HHCLB algorithm. Basically, the Dimension Exchange Method (DEM) approach is used in this paper to construct a new load balancing approach on the network of HHC interconnections. Thus, an algorithm was introduced and simulated using java threads, where the performance of the algorithm is evaluated both analytically and experimentally. The evaluation was in terms of various performance metrics, including, execution time, load balancing accuracy, communication cost. By implementing the proposed load balancing algorithm to the HHC network, a high degree of accuracy and minimal execution time was achieved. It is important to highlight that the algorithm recorded small gap between the execution time for small number of processors and large number of processors. For instance, the algorithm achieved 0.14 seconds for balancing the load of 6 processors while 0.59 seconds for balancing the load of 3072 processors. This proves how effective the algorithm is in balancing the load for different network sizes from small to large number of processors, with a slight difference in execution time.

Dynamic Load Balancing of Cloud Data Center Traffic Based on Software-Defined Networking

Cloud computing has grown rapidly in the last decade, where many tenants are using various cloud applications. Cloud computing users expect and demand to run their applications with the highest performance, reliability, and best quality of service. So, load balancing techniques are essential components and widely used in the cloud data center network to manage users’ requests and distribute applications traffic among available data center resources. Software-defined networking is a highly flexible network architecture that automates network configuration using a centralized controller to overcome the traditional network limitations and manual configuration for every network device. This paper proposes a dynamic load balancing technique using software-defined networking for the cloud data center that can handle different application types in real-time based on traffic type and the required quality of service. The proposed method aims to bridge the gap between application requirements and the resources in the cloud data center by choosing the best path for data transmission and selecting the best performance server to process the users' requests according to traffic type for efficient resources’ utilization, minimizing response time, and maximizing throughput. The simulation results show that the proposed method enhances the performance of the running applications’ and utilizes the data center resources efficiently compared with the current load balancing techniques.

Memory Efficient Load Balancing for Distributed Large-Scale Volume Rendering Using a Two-Layered Group Structure

We propose a novel compositing pipeline and a dynamic load balancing technique for volume rendering which utilizes a two-layered group structure to achieve effective and scalable load balancing. Th ...

A heuristic technique to improve energy efficiency with dynamic load balancing

Heterogeneous computers require a well-distributed workload to operate efficiently. When possible, this load balancing procedure should redistribute the workload with minimal knowledge of the system architecture, to reduce overhead. We propose a generic dynamic load balancing technique for iterative problems, independent from the resource to optimize. Proof of this generalization is given through formalization of the designed technique. A heuristic algorithm is defined based upon this formalization, with a structure that facilitates different objective functions. As a result, swapping the objective function can be done with relatively low effort. This heuristic is implemented to minimize energy consumption in an application. We use this application to solve three different dynamic programming problems with multiple GPUs. The implementation is described and then compared against two different workloads, the homogeneous distribution and another dynamic load balancing technique. Our experimentation shows good results in minimizing the overall energy consumption with low overhead.

Synchronous Q Learning Based Technique for Performance Improvement in Multi core Processors

Conventional processors are widely used in many practical applications such as weather forecasting, AI, Ocean modeling, Big data analysis, etc. In this research work we have investigated various parallel computing approaches to improve the performance in terms of execution time. It is shown by simulation that multi processor systems takes less time. We have literature two algorithms namely Classical Q-learning and Modified Q-learning i.e. Synchronous Q-learning are available for load balancing from this point of view we have developed a new approach of dynamic load balancing technique. In the present work we have combine two algorithms and developed a new algorithms under the name Synchronous Q-learning Algorithms. It is shown by simulation that proposed Synchronous Q-learning algorithms takes less time. In this paper, we exhibit new Synchronous Q learning algorithm that consolidate components of policy iteration and classical Q learning/esteem iteration to effectively learn and control arrangements for a dynamic load adjusting situations utilizing reinforcement learning techniques.

Parallel anisotropic mesh refinement with dynamic load balancing for transonic flow simulations

Abstract The present paper discusses an effective adaptive methods suited for use in parallel environment. An in-house, parallel flow solver based on the residual distribution method is used for the solution of flow problems. Simulation is parallelized based on the domain decomposition approach. Adaptive changes to the mesh are achieved by two distinctive techniques. Mesh refinement is performed by dividing element edges and a subsequent application of pre defined splitting templates. Mesh regularization and derefinement is achieved through topology conserving node movement (r-adaptivity). Parallel implementations of an adaptive use the dynamic load balancing technique.

Performance Analysis of Applying Load Balancing Strategies on Different SDN Environments

Software-Defined Network (SDN) is considered a breakthrough to the global network. It plays an important role in performance improvement and network optimization. SDN is a new mechanism for managing and designing networks rather than the current traditional network system which does not afford more services and higher data rates; therefore, we analyze the effect of applying load balancing techniques and its importance in different SDN environments. In this paper, we propose a dynamic server load balancing technique in SDN architecture. Hence, we implement a server Connection-based load balancing technique and evaluate its performance with a static Round-robin and Random-based in both mininet emulation environment and Raspberry Pi OpenFlow-enabled switch using Ryu OpenFlow controller. The performance of the proposed algorithm is compared with Round-robin and random distribution of clients' requests. The results show that the proposed technique achieves more reliability and higher resource utilization than the Round-robin and Random-based load balancing strategies. In addition, the proposed scheme exhibits more scalability and low-cost characteristics.Software-Defined Network (SDN) is considered a breakthrough to the global network. It plays an important role in performance improvement and network optimization. SDN is a new mechanism for managing and designing networks rather than the current traditional network system which does not afford more services and higher data rates; therefore, we analyze the effect of applying load balancing techniques and its importance in different SDN environments. In this paper, we propose a dynamic server load balancing technique in SDN architecture. Hence, we implement a server Connection-based load balancing technique and evaluate its performance with a static Round-robin and Random-based in both mininet emulation environment and Raspberry Pi OpenFlow-enabled switch using Ryu OpenFlow controller. The performance of the proposed algorithm is compared with Round-robin and random distribution of clients' requests. The results show that the proposed technique achieves more reliability and higher resource utilization than the Round-robin and Random-based load balancing strategies. In addition, the proposed scheme exhibits more scalability and low-cost characteristics.