Load Balancing Scheduling Research Articles

DRALBA: Dynamic and Resource Aware Load Balanced Scheduling Approach for Cloud Computing

For the last few years, Cloud computing has been considered an attractive high-performance computing platform for individuals as well as organizations. The Cloud service providers (CSPs) are setting up data centers with high performance computing resources to accommodate the needs of Cloud users. The users are mainly interested in the response time, whereas the Cloud service providers are more concerned about the revenue generation. Concerning these requirements, the task scheduling for the users’ applications in Cloud computing attained focus from the research community. Various task scheduling heuristics have been proposed that are available in the literature. However, the task scheduling problem is NP-hard in nature and thus finding optimal scheduling is always challenging. In this research, a resource-aware dynamic task scheduling approach is proposed and implemented. The simulation experiments have been performed on the Cloudsim simulation tool considering three renowned datasets, namely HCSP, GoCJ, and Synthetic workload. The obtained results of the proposed approach are then compared against RALBA, Dynamic MaxMin, DLBA, and PSSELB scheduling approaches concerning average resource utilization (ARUR), Makespan, Throughput, and average response time (ART). The DRALBA approach has revealed significant improvements in terms of attained ARUR, Throughput, and Makespan.This fact is endorsed by the average resource utilization results (i.e., 98 % for HCSP dataset, 75 % for Synthetic workload (improve ARUR by 72.00 %, 77.33 %, 78.67 %, and 13.33 % as compared to RALBA, Dynamic MaxMin, DLBA and PSSELB respectively), and 77 % for GoCJ (i.e., the second best attained ARUR)).

A HDFS dynamic load balancing strategy using improved niche PSO algorithm in cloud storage

A Hadoop distributed file system (HDFS) NameNode dynamic load balancing strategy (NDLBT) using improved niche particle swarm optimisation (PSO) is proposed, which is used to solve the load balancing problem of urban surveillance video big data. Firstly, the relationship between bandwidth consumption and the bit rate of video files, the size of data blocks and access hotspots of online video files are analysed. Then, adaptive backup is realised by dynamic multiple copies of heterogeneous nodes. The dynamic distribution of metadata is implemented to ensure the performance of metadata server clusters. Finally, we propose a new improved niche PSO algorithm to achieve load balancing scheduling. In the experimental scenario where high bandwidth consumption video files are used as service access hotspots, the proposed method, which can reduce the bandwidth peak value of bottleneck nodes in data node clusters by 20%, is superior to the original load balancing method in 90% scenarios.

Simulation on dynamic load balancing of distributed parallel computing network system

In order to improve the dynamic load balancing scheduling capability of distributed parallel computing network system and the throughput performance of network, a dynamic load balancing scheduling ...

Construction of load balancing scheduling model for cloud computing task based on chaotic ant colony algorithm

Construction of load balancing scheduling model for cloud computing task based on chaotic ant colony algorithm

AEML: An Acceleration Engine for Multi-GPU Load-balancing in Distributed Heterogeneous Environment

For the rapid growth computation requirements in big data and artificial intelligence area, CPU-GPU heterogeneous clusters can provide more powerful computing capacity compared to CPU clusters. The number of GPUs on single computing node is scalable, which greatly improves the computing capacity of the cluster under the condition of limited cluster size. However, there is a lack of the effective load-balancing scheduling model in multi-GPU hardware environment. This paper proposes AEML, an acceleration engine for multi-GPU load-balancing in distributed heterogeneous environment. AEML can effectively integrate GPUs into distributed processing framework and achieve great load-balance among multiple heterogeneous GPUs. We propose a heterogeneous task execution model based on multiple GPUs and multiple streams (MGMS), which can effectively balance the workload of multiple GPUs. MGMS model utilizes four core techniques: a fine-grained task mapping mechanism, a device resource unified management scheme, a novel resource-aware GPU task scheduling strategy and a feedback-based streams adjustment scheme. The implementation of AEML system is based on Spark 2.4.1 and NVIDIA CUDA 10.0. We comprehensively evaluate the performance of AEML with multiple typical benchmarks. Experimental results show that AEML can fully exploit the computing power of GPUs and achieve great load-balance among multiple heterogeneous GPUs.

Optimizing Data Pipeline Performance in Modern GPU Architectures

Optimizing data pipeline performance in modern GPU architectures is critical for achieving high computational throughput and efficient resource utilization in data-intensive applications. With the rise of deep learning, scientific simulations, and real-time analytics, GPUs have become integral in accelerating data processing tasks. However, ensuring optimal performance involves addressing several challenges, such as memory bandwidth limitations, data transfer bottlenecks between CPU and GPU, and efficient parallel execution of workloads. This research explores techniques for improving data pipeline performance by focusing on memory management, load balancing, and task scheduling. One key strategy is optimizing data movement through techniques like memory coalescing, which minimizes access latency, and overlapping data transfers with computation. Furthermore, leveraging the architectural advances in modern GPUs, such as unified memory and NVLink, can significantly reduce data transfer overhead. Task parallelism and efficient workload distribution across multiple GPU cores also play a crucial role in enhancing pipeline throughput. Additionally, the study highlights the importance of tuning GPU kernels and optimizing data preprocessing steps to ensure minimal latency and maximum throughput. By adopting advanced profiling tools and performance monitoring techniques, bottlenecks can be identified, and pipeline optimization strategies can be fine-tuned. The findings presented provide a comprehensive approach for designing and optimizing data pipelines, leading to significant performance improvements in GPU-based systems, ultimately driving the next generation of high-performance computing applications.

DDMTS: A novel dynamic load balancing scheduling scheme under SLA constraints in cloud computing

Cloud computing is a computing method based on the Internet designed to share resources through virtualization technology. For a large number of requests waiting to be processed, task scheduling is used to reasonably allocate computing resources to requests. With the rapid development of computer hardware and software, deep reinforcement learning (DRL) provides a new direction for better solving task scheduling problems. In this paper, we propose a novel DRL-based dynamic load balancing task scheduling algorithm under service-level agreement (SLA) constraints to reduce the load imbalance of virtual machines (VMs) and task rejection rate. First, we use the DRL method to select a suitable VM for the task and then determine whether to execute the task on the selected VM violates the SLA. If the SLA is violated, the task is refused and feedback a negative reward for DRL training; otherwise, the task is received and executed, and feedback a reward according to the balance of the VMs load after the task is executed. Compared with three other task scheduling algorithms applied to randomly generated benchmark and Google real user workload trace benchmark, the proposed algorithm exhibits the best performance in balancing VMs load and reducing the task rejection rate, improving the overall level of cloud computing services.

Efficient load balancing techniques for multi-datacenter cloud milieu

Load balancing is a challenge in the cloud network for proper resource utilization and improvement of makespan. To overcome the problem of load balancing various static and dynamic scheduling methods come into existence. In this paper, we have proposed two multi-datacenter two-phase load adjustment techniques for load balancing and apposite task scheduling. Our proposed methods allocate user jobs to different virtual machines present in different datacenters based on better makespan. Here we have considered communication delay and bandwidth requirement for inter-datacenter task migration. The simulation result specifies that proposed techniques give a significant improvement in makespan and resource utilization when compared with existing algorithms.

Fault‐tolerant with load balancing scheduling in a fog‐based IoT application

Fog computing (FC) with a distributed architecture plays an essential role in Internet-of-Things (IoT). This paradigm utilises the processing abilities of Fog devices (FDs) and decreases latency. The large volume of data and its process in IoT can cause network failures. Researchers tend to consider communication reliability to reduce fault effects and achieve high performance. Fault tolerance becomes a necessary matter to enhance the reliability of the Fog. Notably, fault tolerance studies have been performed mostly on the Cloud system. To counter this issue, the authors propose a novel fault-tolerant scheduling algorithm of modules in FC and optimise it. The main idea of this approach is a classification method for different modules alongside of computing the energy consumption of all FDs and finding minimal FDs' energy consumption. To distribute modules between FDs, they present an energy-efficient checkpointing and load balancing technique based on the Bayesian classification and call it by ECLB. The performance of the proposed method is evaluated by comparing it with the state-of-the-art algorithms in terms of delay, energy consumption, execution cost, network usage, and total executed modules. Analysis and simulation results indicate that the authors' methods are efficient and superior to others.

An Optimization on Task Scheduling for Makespan, Energy Consumption, and Load Balancing in Cloud Computing Using Meta-Heuristic

An Optimization on Task Scheduling for Makespan, Energy Consumption, and Load Balancing in Cloud Computing Using Meta-Heuristic

RETRACTED ARTICLE: SLA-aware load balancing using risk management framework in cloud

In cloud computing environment, load balancing is an important task. So many of the researchers had focused on load balanced scheduling technique. Those provides better load balancing in cloud but there are some issues like resource allocation and cost maintenance. One of the major issue in load balancing techniques is service level agreement (SLA) management because many of them are affected by this SLA-violation. Many researchers have proposed various risk based framework but few of them has guides the service provider to take steps for SLA violation abatement and they also need some improvements. To tackle this problem, a new SLA-aware risk management framework (SA-RMF) is proposed in this work for efficient load balancing in cloud. A new technique is presented here based on CPU parameter for generating efficient dynamic threshold. A better quality of service values prediction is achieved by using hybrid approach in SA-RMF. Through experiments, the appropriateness of proposed system is demonstrated and validated that it helps cloud providers to mitigate future violations of services and consequences.

Hybrid Load-Balanced Scheduling in Scalable Cloud Environment

Cloud computing is a high computational distributed environment with high reliability and quality of service. It is playing an important role in the next generation of computing with pay per use model and high elasticity. With increased requirement for cloud resources, load over the cloud servers has increased, which makes cloud use a more efficient algorithm to maintain its performance and quality of service to users. The performance metrics that define the performance of task scheduling include execution time, finish time, scheduling time, task completion cost, and load balancing on each computing resources. So, to overcome existing solutions and provide better QoS performance, a neural-network-based GA-ANN scheduling algorithm is proposed in this paper, which outperforms the existing solutions. To simulate the proposed GA-ANN model, cloudsim3.0 toolkit is used, and the performance is evaluated by comparing simulation time, average start time, average finish time, execution time, and utilization percentage of computing resources (VMs).

Load-balancing scheduling of simulation tasks based on a static-dynamic hybrid algorithm

ABSTRACT A scheduling algorithm is crucial for running a simulation model so that tasks can be performed efficiently.The traditionally used blade-based parallel engine system cannot be adapted to a new simulation model. This study proposed a combined dynamic priority and static method, that is, a hybrid load-balancing scheduling (HLB) algorithm. The algorithm is given priority according to the operating cycle of the model and system steps. The experimental results demonstrated that the algorithm outperformed the earliest deadline first and the time-stepped load-balancing scheduling algorithms. The results also demonstrated that the HLBhad a higher real-time operating efficiency than the other algorithms under a lower overhead guarantee. The HLB algorithm causedbetter performance while maintaining computation and communication efficiency. Simultaneously, the utilisation rate of the central processing unit was around 35%.The further study should be enhanced to generalise it so that it could be applied to incorporate load balancing.

A cost saving and load balancing task scheduling model for computational biology in heterogeneous cloud datacenters

Cloud-based scientific workflow systems can play an important role in the development of cost-effective bioinformatics analysis applications. There are differences in the cost control and performance of many kinds of servers in heterogeneous cloud data centers for bioinformatics workflows running, which can lead to imbalance between operational/maintenance management costs and quality of service of server clusters. A task scheduling model that responds to the peaks and valleys of task sequencing—the number of tasks that arrive in a given unit of time—is related to indicators such as cost saving, load balancing and system performance (average task wait time, average response time and throughput). This study proposes a large-scale cost-saving and load-balancing scheduling model, called HDCBS, for the optimization of system throughput. First, queuing theory is used to model each computing node as an independent queuing system and to obtain the average system wait time and average task response time. Then, using convex optimization theory, a task assignment solution is proposed with a load-balancing mechanism. The validity of the task scheduling model is verified by simulation experiments, and the model performance is further validated through a comparison with other frequently used scheduling methods. The simulation results show that the credibility of HDCBS is greater than 95% in task scheduling.

RETRACTED ARTICLE: Secure storage allocation scheme using fuzzy based heuristic algorithm for cloud

Cloud computing is a paradigm in the modern era for the development of secure storage and secure job scheduling process in a cloud-based system. The cloud technology provides a user interface in the sense of raw data management with the relational databases to improve flexibility and scalability. Fuzzy-based Heuristic algorithm is proposed for secure storage allocation in a cloud environment. In this work, client systems are clustered using a fuzzy-based rule mechanism. The server generates a distributed public key using RSA for ensuring security and resolve memory recycling issues. It enhances the usage of the client over the cloud data storage in secure manner. It also improves the efficiency of data storage globally by implementing heuristic techniques. The K-nearest neighbors approach is used for efficient query searching for data storage in the clusters by the client. The proposed load-balanced scheduling approach enhances power consumption and efficiency. In the simulation, the results reveal that the proposed scheme provides better performance than existing approaches in terms of load balancing with security.

An Algorithmic Analysis on Load Balancing and Workflow Scheduling in Cloud

Cloud computing is a paradigm for facilitating omnipresent, comfortable, on-demand network access to shared provisions of configurable computing resources like servers, applications, networks, and services that can be quickly provisioned and discharged with minimal management effort or service provider intercommunication. Data centers are the set of hosts that runs Virtual Machines. The broker submits the requests which are provisioned to the data centers that submit the task. The task or an application that runs is known as cloudlet. The term “Cloud computing” is to deliver the hosted services on the internet. Each “cloudlet” requires one or more computing resources from the cloud to complete its job. The executions of these sub-tasks use a scheduling algorithm in a particular fashion. Hence, one of the most important issues is that executing a large number of workflow in cloud environment and consumption of resources. The work proposes literature about the scheduling algorithm that resolves the conflict in the allocation of resources and manages the resource provisioning at dynamic to each task in any workflow. This paper proposes an analysis of proposed schemes with different workflow by executing the scheduling algorithm for the cloud workflow. This paper is presented with their objectives, parameters, and limitations and finally concluded with research directions.

An efficient load balancing system using adaptive dragonfly algorithm in cloud computing

With the rapid development of processing and storage technologies and the success of the Internet, computing resources have become cheaper, more powerful and more ubiquitously available than ever before. This technological trend has enabled the realization of a new computing model, called cloud computing. In cloud, scheduling is an important application. In cloud environments, load balancing task scheduling is an important problem that directly affects resource utilization. Undoubtedly, load balancing scheduling is a serious aspect that should be considered because of its significant impact on both the back end and the front end of the cloud research industry. Good resource utilization is achieved whenever an effective load balance is achieved in the cloud. But, load balancing in cloud computing is an NP-hard optimization problem. In order to accomplish this problem, a novel load balancing task scheduling algorithm in cloud using Adaptive Dragonfly algorithm (ADA) is proposed. The ADA is a combination of dragonfly algorithm and firefly algorithm. Moreover, to attain the better performance, multi-objective function is developed based on three parameters namely, completion time, processing costs and load. Finally, the performance of proposed methodology is evaluated in terms of different metrics namely, execution cost and execution time. The experimental results demonstrate that a proposed approach accomplishes better load balancing result compared to other approaches.

Load Balancing Scheduling of Power Network in Cloud Computing Environment

In the traditional cloud computing environment, the power network load balancing scheduling method cannot adaptively track the time-varying coupling characteristics of load data, and the power data center is not efficient in processing massive data, so it is increasingly unable to meet the multi-service quality requirements of power users. The purpose of this paper is to improve the load balancing of power network, improve the efficiency of power network balance scheduling and power resource balance allocation in cloud computing environment. At the research method level, this paper analyzes the application status of data center and cloud computing task scheduling, and starts from the task scheduling demand analysis of cloud computing data center to determine load balancing and QoS as the technical goals of task scheduling. According to the parallel model framework of MapReduce, the task execution process is studied, and the design scheme of power cloud data center is given in view of the problems of low resource utilization, poor scalability and high energy consumption cost faced by the existing power data center. The experimental results show that the network load balancing scheduling method proposed in this paper can meet the multi-qos demand of power users to the maximum extent, effectively improve the efficiency of power data center operation, and achieve better load balancing effect than the traditional scheme.

Live Virtual Machine migration: A Performance Test Xen and KVM

Virtualization technology has many important features such as live virtual machine migration. In live virtual machine migration, a power on virtual machine is moved from one physical host to another. It has various benefits such as server consolidation, proactive failure, load balancing, energy saving and resource scheduling. Live virtual machine migration is very useful tool in cluster environment, administrators of data centers and in cloud environment. Live virtual machine migration is supported by hypervisors such as Xen, KVM, VMware etc. In this paper we discuss live virtual machine migration Pre-Copy approach which is a default approach in many hypervisors. We compare the performance of virtual machines which are made using Xen and KVM. We also compare performance when virtual machines are migrate using Xen and KVM in cloudreport simulator. In result we find that KVM performs better than Xen.

Software defined network using enhanced workflow scheduling in surveillance

In recent years, with the rapid development of information transfer technology, the load balancing scheduling of server cluster system must be optimized. In order to optimize the load of server, a load balancing scheduling strategy for server cluster system was proposed in this paper. A new scheduling weighted method was proposed, and the optimization of load allocation was realized. In order to verify the feasibility of the scheduling strategy, the results were tested from several aspects. Through a comprehensive test, the research of load balancing scheduling strategy optimization for server cluster system is feasible, which can further improve the server’s load capacity.