Load Balancing Evaluation Tools for a Private Cloud: A Comparative Study

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.

Load balancing is an important concept which helps to raise the throughput. It helps to increase the performance of the system by assigning virtual machine for the execution of user request in less time. It assists in raising user satisfaction level and reducing user response time. As we know, the demand for cloud services raises day by day, which lead to load balancing as a major problem. The technique helps to distribute the workload among all available servers so that user can get their resource in less time. Cloud load balancing introduced the distribution of workload traffic and demands that reside over the Internet. It takes advantage of the cloud extensibility and possesses alertness to meet redirected workload demands and to improve long-term opportunity. The purpose of load balancing is to continue the system constancy and develop the performance considerably. We propose a load balancing technique by using c-means clustering to handle tasks efficiently which conserve less amount of energy.

Goal: Load balancing policies often map workloads on virtual machines, and are being sought to achieve their goals by creating an almost equal level of workload on any virtual machine. In this research, a hybrid load balancing algorithm is proposed with the aim of reducing response time and processing time.
 Design / Methodology / Approach: The proposed algorithm performs load balancing using a table including the status indicators of virtual machines and the task list allocated to each virtual machine. The evaluation results of response time and processing time in data centers from four algorithms, ESCE, Throttled, Round Robin and the proposed algorithm is done.
 Results: The overall response time and data processing time in the proposed algorithm data center are shorter than other algorithms and improve the response time and data processing time in the data center. The results of the overall response time for all algorithms show that the response time of the proposed algorithm is 12.28%, compared to the Round Robin algorithm, 9.1% compared to the Throttled algorithm, and 4.86% of the ESCE algorithm.
 Limitations of the investigation: Due to time and technical limitations, load balancing has not been achieved with more goals, such as lowering costs and increasing productivity.
 Practical implications: The implementation of a hybrid load factor policy can improve the response time and processing time. The use of load balancing will cause the traffic load between virtual machines to be properly distributed and prevent bottlenecks. This will be effective in increasing customer responsiveness. And finally, improving response time increases the satisfaction of cloud users and increases the productivity of computing resources.
 Originality/Value: This research can be effective in optimizing the existing algorithms and will take a step towards further research in this regard.

With the development of economics and the globalization of trade, the Multimedia Conference System is now widely used and the number of its users increases exponentially. But a single server can't handle the high concurrent user requests effectively. In this paper, we improved the Multimedia Conference System with load balance technology. With the load balance technology, the robustness and availability of the system can be enhanced significantly. Keywords-load balance, cluster, Nginx, Multimedia Conference System I. INTRODUCTION With the explosive popularity of the Internet and the World Wide Web, most popular web sites confronted with a problem that the server overloaded caused by the increasing number of concurrent accesses. In order to process the user requests timely, increase the network throughput and improve the quality of service, it's necessary to upgrade the hardware and software of the sever. But the the hardware is expensive and non-scalability. In this case, server cluster system appears. The sever cluster system refers to a server group which is composed of more than one homogeneous or heterogeneous severs and can provides services that are transparent for the external users. And it becomes a key point that how to achieve a reasonable distribution of load between multiple servers and avoid appearing one server with full load while other servers with little load. In this situation, the load balance technology was born. In recent years, cluster technology and load balance technology got fully developed and they have significant effect on solving the problem of overloading. As a web service, the Multimedia Conference System also has to face the problem that the overload of the server and uneven distribution of resources when there are too many users access the system. Using cluster and load balance technology, the requests the system received can be evenly distributed on each server, thus we can reduce the response time of the requests and improve the resource utilization. The Multimedia Conference System proposed in this paper contains Client Layer, Load Balancer Layer, Server Cluster Layer and Media Server Layer. The Client Layer receives the requests of the user including creating a conference and joining an existing conference. The Load Balancer Layer redirect the user's request to one of the application servers in the cluster according to the load balance algorithm and the different parameters from the Client Layer. The application servers in Server Cluster Layer call the Media Server resources and return them to the Client Layer. This implements a basic process of the system. The Load Balancer Layer contains four modules including User Authentication, Decision-Making, Redirecting and Log Parsing. With the Load Balancer Layer and its load balance algorithm, the user's request can be distributed on different servers and those who want to join the same conference can be allocated to the same application server. Thus we can minimize the response time of the request, improve the resource utilization and also ensure the conference can be held correctly. The rest of this paper is organized as follows: we describe related work in section II. Section III presents the architecture of the Multimedia Conference System. In section IV, we display the implementation of the Load Balancer Layer in detail. Some experiments of key features of load balancer in section V. We conclude in section VI.

ABSTRACTOver the past decade, as demand for cloud services has surged, the strategic selection of these services has become increasingly crucial. The growing complexity within the cloud industry underscores the urgent need for a robust model for choosing cloud services effectively. Users often struggle to make informed decisions due to the dynamic nature and varying quality of available cloud services. In response, this paper introduces a novel decision‐making approach aimed at optimizing the selection process by identifying the most suitable combination of cloud services. The focus is on integrating these services into a cohesive ensemble to better fulfill user requirements. In contrast to existing methodologies, our approach evaluates cloud services on a continuous scale, taking into account critical tasks such as workload balancing, storage management, and network resource handling. We propose a model for selecting optimal composite cloud services, which includes real‐time optimization and addresses the consideration of null values for Quality of Service (QoS)‐based attributes (e.g., response time, cost, availability, and reliability) in the dataset—a factor overlooked by current literature. The proposed algorithm is inspired by computational intelligence and driven by an evolutionary algorithm‐based approach that undergoes evaluation across multiple datasets. The results illustrate its superiority, showcasing its ability to outperform existing optimization‐based methods in terms of execution time.

Numerous works has been done for load balancing of web servers in grid environment. Reason behinds popularity of grid environment is to allow accessing distributed resources which are located at remote locations. For effective utilization, load must be balanced among all resources. Importance of load balancing is discussed by distinguishing the system between without load balancing and with load balancing. Various performance metrics that need to be considered for designing an efficient load balancing algorithm are also described. Intensive review of literature of different load balancing approaches for web servers have been carried out and presented in this paper. On the basis of literature survey, a new dynamic load balancing mechanism is proposed for optimizing waiting time and cost.

A Uniform Framework for Dynamic Load Balancing Strategies in Distributed Processing Systems

Web server load balancing: A queueing analysis

<p>The distribution of scheduler from user inquiries in the clouds is complex. In keeping up with the cloud computing environment and the inquirers, the clouds meet with some problematic load balancing complications as an improving load balancing tool induces the rigorous efficiency of the cloud based website’s user access. Overloaded or underloaded conditions originate processing catastrophe regarding the prolonged execution time, bandwidth hog, malfunction, and etc. Besides, to manipulate Erlang concurrent tasks is another skyward situation. Hence, the load balancing is obliged to exhaust all mentioned conditions. The proposed load balancing algorithm for Erlang concurrent tasks (those are and could also be autonomous and unstable.) on VMware workstations is introduced. There are several load patterns within the clouds corresponding to CPU’s load (utilization), memory load (queue size), link capacity load (bandwidth), and so on. The proposed load balancing is to spot underloaded and overloaded conditions then stabilizes the weight amidst computing nodes. There are countless load balancing approaches in the cloud environment to examine performance parameters. A short outline of corresponding performance metrics in the review and their findings are presented. To investigate the fit efficiency of the proposed algorithm, the simulation is applied then results based on the proposed method are compared to the existing ones. The outcomes settle the weight balancing, outperform others when executing Erlang traffic, and are catered in the context.</p>

A load balancing strategy for software defined networking (SDN) in the cloud seems to resolve the availability and latency difficulties in the HTTP server nowadays, thanks to the strong demand for cloud services. In this article, we show how to implement a load balancing system for software defined networking (SDN) in the cloud. The HTTP server handled 6 million requests, with 400 requests per second. Due to overload, the website's server will not reply to any more requests sent by the client. hence, after applying load balance, the same parameters may be measured using a load balancing system. The 6 million requests, each involving four connections, were all successfully accepted by the cloud environment's HA proxy, allowing the traffic from the HTTP server to be dispersed across a pool of servers rather than a single server. Having an HTTP load balancer helps to improve the availability and latency of HTTP requests, and the HA proxy eventually gives an adequate answer time with no requests lost, even after 6 million requests arriving at the same time.

Cloud computing is a new era technology, which is entirely dependent on the internet to maintain large applications, where data is shared over one platform to provide better services to clients belonging to different organizations. It ensures maximum utilization of computational resources by making availability of data, software and infrastructure with at a lower cost in a secure, reliable and flexible manner. Fog computing is emerging as a powerful and popular computing paradigm, which extends cloud computing paradigm to enable the service execution in the edge network. The mobile and IoT (Internet of Things) applications could choose the computing nodes in both fog and cloud for resource provisioning. Generally, load balancing is one of the key factors to achieve resource efficiency and avoid bottleneck, overloaded and low-load resource usage. Load balancing is an important aspect of fog networks that avoids a situation with some under-loaded or overloaded fog nodes. Quality of Service (QoS) parameters such as resource utilization, cost, response time, performance, and energy consumption can be improved with load balancing. Recently, several studies have been conducted on load balancing in fog computing. This paper introduces an Artificial Immune System (AIS)-based load balancer for efficient resource allocation in fog computing, utilizing a multi-metric fitness function incorporating CPU and RAM availability, energy efficiency, and proximity. A mutation mechanism enhances adaptability and prevents allocation bias. the results reveal that the AIS achieves 41.67% better resource allocation efficiency compared to Random Load Balancer and up to 30.77% improvement over Least connection and Round Robin strategies. The AIS model effectively balances CPU utilization, latency, and energy consumption, providing a robust and scalable solution for dynamic fog networks.

Cloud computing is a recent computer technology that has virtualized infrastructure to offer secure and reliable services to users in a complex environment. The main advantage of the cloud is its scalability. Vast scalability is possible because of load balancing. Due to the advancement of the cloud, huge numbers of service provisioning requests are generated from users. Cloud needs a better and more efficient load-balancing mechanism to handle all the user’s requests. This paper proposes the principal component gradient round robin load balancing (PCGRLB) technique to balance the workload in a cloud server for handling a huge workload within a minimum time. The PCGRLB technique performs two processes: Virtual machine analysis and load balancing. The PCGRLB technique is evaluated on factors such as throughput, makespan, and response time concerning several tasks. The results show that the PCGRLB technique offers an efficient solution in achieving higher load balance, efficiency, throughput, and minimized makespan and response time than the conventional techniques.

The increasing complexity of web applications and the growing number of concurrent users require load-balancing mechanisms that can maintain system performance and service reliability. This study aims to evaluate and compare the performance of NGINX and Caddy as load balancers for WordPress applications deployed in a Podman-based containerized environment. A quantitative experimental approach is employed by measuring key Quality of Service (QoS) metrics, including request rate, response time, and error rate, across multiple load scenarios and load-balancing algorithms, namely round-robin and least connection. The experimental results indicate that under low to medium load conditions, both NGINX and Caddy exhibit comparable performance. However, under high-load conditions, performance differences become significant. Caddy, particularly when using the least-connection algorithm, can maintain substantially higher throughput, achieving an aggregate request rate of approximately 2349.55 req/s, compared to NGINX, which records approximately 426.02 req/s under comparable configurations. In terms of response time, the P90 response time for Caddy is approximately 289.79 ms, whereas NGINX exceeds 4 seconds. Furthermore, Caddy demonstrates superior reliability with an error rate of 0% across all testing scenarios, while NGINX experiences extremely high request failure rates under extreme load conditions. These findings suggest that Caddy, particularly when combined with the least-connection algorithm, provides a more adaptive and stable load-balancing solution for dynamic, containerized web applications with fluctuating traffic patterns. Nevertheless, this study is conducted in a controlled environment using baseline configurations without advanced optimizations; therefore, the results represent a relative performance comparison rather than an absolute benchmark for all production environments. Future research is recommended to investigate advanced configurations, different web application types, and more diverse workload scenarios to achieve a more comprehensive understanding of load balancer performance in containerized environments.

In the past few years, cloud computing has emerged as a vigorous technology in the era of internet. Its economic model fecilitate the client to use software, infrastructure, applications as a service by just paying according to the usage without possessing them. The development of cloud computing needs to balance the load among different replicated services offered by cloud environment. In terms of responsiveness and availability, load balancing is one major aspect to deal with. Load balancing ensures that the load is equally spread over the cloud servers. Efficient load balancing algorithm leads to decrease in maintenance costs and increase in resource availability. There are various factors that affect the efficiency of load balancing algorithm, two of such factors are addressed in the proposed algorithm. The download of large data files is handled by dual direction technique. This technique make use of replicated servers and download the partitions of a file simultaneously from multiple replica servers. Another factor that is taken into consideration is handling of high priority tasks. There are various parameters to define the effectiveness of a load balancing algorithm such as response time, data center processing time etc. In this present work a load balancing algorithm, based on honey bee optimization algorithm is proposed which has shown better response time. Experimental studies on standard data sets showed that the proposed request balanced honey bee algorithm compared to the original honey bee algorithm, has performed 5.2% to 9.7% better with respect to the response time and 13.8% to 68.2% better with respect to the processing time, which is a significant improvement.

In parallel I/O systems, fast response to disk access and load balancing are two important performance objectives pursued by end users and applications. These performances are largely determined by the data allocation strategies, or file assignment algorithms. However, most existing algorithms can only obtain one of the performance objectives, including the well-known algorithms like Greedy, Sort Partition (SP) and Hybrid Partition (HP). New algorithms that can achieve both objectives are very necessary for parallel I/O systems. In this paper, we have proposed two new allocation algorithms for file assignment in parallel I/O systems: an offline Balanced Allocation with Sort (BAS) algorithm and an online Balanced Allocation with Sort for Batch (BASB) algorithm. Both algorithms aim to achieve the optimal mean response time and load balancing at the same time. The experiment results have shown that the BAS algorithm can get the optimal performance on response time among all compared algorithms and better performance on load balancing than SP. For online BASB algorithm, it can achieve the optimal performances on both response time and load balancing among all algorithms for comparison.