Server Load Balancing Research Articles

A Literature Survey on Algorithms for the Optimal Load Balancing in Cloud Computing Environments

Abstract: This paper provides a survey of the existing literature and research carried out in the area of Optimal load balancing in servers. Load balancing in cloud computing is one of the important aspects for efficient delivery of resources and computing . The process of distributing workloads and computing resources in a cloud computing environment is cloud load balancing. It allows enterprises to manage application or workload demands by allocating resources among multiple computers, networks or servers. Hosting the distribution of workload traffic and demands that reside over the internet. It helps enterprises achieve high performance levels for very reasonable costs which is lower than traditional on premises load balancing technology . By taking advantage of clouds' scalability and agility cloud load balancing technology meets rerouted work demands and also improves overall availability. Load balancing can provide health checks for cloud in addition to the workload and traffic distribution .Performance analysis of different existing load balancing algorithms based on different parameters is to be carried out and the algorithm will be optimized for the better performance in cloud. The main purpose of the proposed paper will be to get help in the design of new algorithms in future.

Correction to: Deep Reinforcement Learning for Load Balancing of Edge Servers in IoV

Correction to: Deep Reinforcement Learning for Load Balancing of Edge Servers in IoV

A two-stage container management in the cloud for optimizing the load balancing and migration cost

Container technology is driving the transformation of Container as a Service (CaaS), which calls for an optimal solution of container management. In this paper, we present a two-stage framework of container management in the cloud, which includes container placement and container migration. Specifically, we investigate the container placement strategy of cloud servers for load balancing in a long time scale. In a short time scale, we investigate the container migration strategy among the cloud servers for minimizing the migration cost while maintaining the degree of load balance. Both of the strategies are formulated as constrained optimization problems. We propose Balance Aware Container Placement (BACP) algorithm and Adaptive Threshold Container Migration (ATCM) algorithm to solve the optimization problems in different time scales. Furthermore, workload prediction of the containers is proposed to enhance the solution. We show by experiments that we can achieve the load balance while reducing the migration cost. Moreover, the proposed algorithm outperforms the meta-heuristic algorithms, i.e., PSO (Particle Swarm Optimization) and CRO (Coral Reef Optimization).

Design of Personalised English Distance Teaching Platform Based on Artificial Intelligence

In the application of the existing distance education platform, the server load is often unbalanced, which increases the response time of the platform. This paper designs a personalised English distance teaching platform based on artificial intelligence. Put forward the overall architecture of the platform, and design the platform on this basis. In order to realise the personalised course resource recommendation function, a collaborative filtering personalised recommendation algorithm is designed to get the best course recommendation results. Using artificial intelligence technology to set up multipoint control unit, the task balance is allocated to multiple operation units, and the reasonable allocation of curriculum resources is realised. According to the functional requirements of the platform design, the composition of functional modules is further set to realise the scheduling and management of curriculum resources. The experimental results show that, under the condition of the same number of concurrent users, the average response time of this design platform is less than that of the existing distance teaching platform, which shows that it has certain advantages in server load balancing and resource scheduling, which can improve the rapid response ability of the platform and enhance the stability of the platform.

Server Load Balancing Analysis for Client Task Assignment in Distributed Systems

The client-server model is widely used in many distributed systems in which proper assignment of client is an important factor for the better performance of system. A large number of clients may exist in a distributed system that establishes communication with one another through a variety of intermediary servers. The primary objective of the overall system is to allocate clients to servers to maintain the load balancing as well as communication cost. There are two criteria for server load evaluation. The first one is complete load, followed by task scheduling, with complete load and task scheduling being diametrically opposed measurements. It seems to be NP-hard to figure out the appropriate client-server allotment for a distributed system's total load and load balancing. In this research, we examine distinct server load balancing strategies for the assignment of clients in distributed systems.

Towards Sustainable Edge Computing Through Renewable Energy Resources and Online, Distributed and Predictive Scheduling

In this work, we tackle the energy consumption problem of edge computing technology looking at two key aspects: (i) reducing the energy burden of modern edge computing facilities to the power grid and (ii) distributing the user-generated computing load within the edge while meeting computing deadlines and achieving network level benefits (server load balancing <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">vs</i> consolidation and reduction of transmission costs). In the considered setup, edge servers are co-located with the base stations of a mobile network. Renewable energy sources are available to power base stations and servers, and users generate workload that is to be processed within certain deadlines. We propose a predictive, online and distributed algorithm for the scheduling of computing jobs that attains objectives (i) and (ii). The algorithm achieves fast convergence, leading to an energy efficient use of edge computing facilities, and obtains in the best case a reduction of 50% in the amount of renewable energy that is sold to the power grid by heuristic policies and that is, in turn, used at the network edge for processing.

Discrete-event simulation of load distribution between server stations

The simulation model of the computational load balancing in a server complex using a balancing server is proposed. The server load balancing model was developed using MATLAB/Simulink/SimEvents/Stateflow tools. The SimEvents based model makes it possible to simulate the server complex as a queuing system and to take into account the sporadic occurrence of requests. The Stateflow based event-driven model simulates the system with a variable time delay in data transmission channels and simulates state transitions for servers. This discrete-event simulation model allows evaluating the possibility of rational use of computational resources and reducing the time of service. In addition, the model allows to investigate the effectiveness of load balancing systems at the stage of their development.

Load Balancing Based on Closed Loop Control Theory (LBBCLCT): A Software Defined Networking (SDN) powered server load balancing system based on closed loop control theory

AbstractThis article comprises work concerned with proposing a scheme for intercepting network traffic and directs that traffic to servers in Software defined network model. Several studies have explored the effects of average response time of user requests with different load balancing algorithms. Server load balancing is a well‐recognized issue which necessitates a better approach by doing deeper understanding of the network parameters. The implications of the mean response time of client requests are investigated in this article by having a closer look on changing the important network parameter namely probing time (time to investigate the server farm in order to gain access to server data) using the concept of closed loop control theory. Our LBBCLCT proves to be effectively distributes the user requests to the most appropriate server by timely changing the server even when the traffic exhibits inconsistency. The results reveal that LBBCLCT outperforms other methods such as round robin, LBBSRT, and SD‐WLB in terms of processing the client requests with a reduced mean server response time. LBBCLCT showed the improvement of average response time by 46.2%, 25.6%, and 43.59% over round robin, LBBSRT, and SDWLB respectively.

Effectively Handling Network Congestion and Load Balancing in Software-Defined Networking

The concept of Software-Defined Networking (SDN) evolves to overcome the drawbacks of the traditional networks with Internet Protocol (I.P.) packets sending and packets handling. The SDN structure is one of the critical advantages of efficiently separating the data plane from the control plane to manage the network configurations and network management. Whenever there are multiple sending devices inside the SDN network, the OpenFlow switches are programmed to handle the limited number of requests for their interface. When the recommendations are exceeded from the specific threshold, the load on the switches also increases. This research article introduces a new approach named LBoBS to handle load balancing by adding the load balancing server to the SDN network. Besides, it is used to maximize SDN’s reliability and efficiency. It also works in coordination with the controller to effectively handle the load balancing policies. The load balancing server is implemented to manage the switches load effectively. Results are evaluated on the NS-3 simulator for packet delivery, bandwidth utilization, latency control, and packet decision ratios on the OpenFlow switches. It has been found that the proposed method improved SDN’s load balancing by 70% compared to the previous state-of-the-art methods.

Adaptive Server Load Balancing in SDN Using PID Neural Network Controller

Web service applications are increasing tremendously in support of high-level businesses. There must be a need of better server load balancing mechanism for improving the performance of web services in business. Though many load balancing methods exist, there is still a need for sophisticated load balancing mechanism for not letting the clients to get frustrated. In this work, the server with minimum response time and the server having less traffic volume were selected for the aimed server to process the forthcoming requests. The Servers are probed with adaptive control of time with two thresholds L and U to indicate the status of server load in terms of response time difference as low, medium and high load by the load balancing application. Fetching the real time responses of entire servers in the server farm is a key component of this intelligent Load balancing system. Many Load Balancing schemes are based on the graded thresholds, because the exact information about the network flux is difficult to obtain. Using two thresholds L and U, it is possible to indicate the load on particular server as low, medium or high depending on the Maximum response time difference of the servers present in the server farm which is below L, between L and U or above U respectively. However, the existing works of load balancing in the server farm incorporate fixed time to measure real time response time, which in general are not optimal for all traffic conditions. Therefore, an algorithm based on Proportional Integration and Derivative neural network controller was designed with two thresholds for tuning the timing to probe the server for near optimal performance. The emulation results has shown a significant gain in the performance by tuning the threshold time. In addition to that, tuning algorithm is implemented in conjunction with Load Balancing scheme which does not tune the fixed time slots.

Design and Implementation of Ship Water Pollution Supervision and Service Information System

In order to improve the ship pollutant discharge monitoring system, control the ship pollutant discharge, establish a more convenient and efficient information communication channel for ship pollutant discharge and receiving services, a ship water pollution discharge system is designed and implemented. This system is divided into three workbenches, namely ship workbenches, enterprise workbenches and supervision workbenches. It is developed based on the front-end and back-end separation mode. The front-end uses Vue framework to implement page function, and Element UI to complete page UI design, Mapbox to display maps, and ECharts component to draw charts. The back-end is implemented through the Spring Boot framework. Redis is used for data caching to improve concurrency and provide data interfaces. Spring Security + JWT is used to achieve single sign on and permission control. The Nginx service gateway is used to achieve service load balancing. MySQL database is used to storage the water pollution discharge information. AIS historical data storage is realized through the distributed database system of Doris cluster.

Improved superheat control of variable-speed vapor compression systems in provision of fast load balancing services

Utilization of renewable energy resources, such as solar and wind, has seen rapid growth on the power grid worldwide. To mitigate the adverse impact of these volatile generation on the reliability of the power grid, fast balancing services are increasingly procured by system operators from various resources, such as flexible air-conditioning and space heating loads. Recent work of the authors has demonstrated technical feasibility and various benefits of using variable-speed vapor compression equipment for two fast load balancing services, namely frequency regulation and solar photovoltaic power smoothing. A major control issue was also identified in prior experimental tests: the drastic changes of the compressor speed for fast load balancing have caused severe superheat regulation issues including oscillations and even wet compression. To this end, a gain-scheduled feedforward and proportional integral (PI) controller is proposed and presented in this paper, which uses the real-time compressor speed in determining compensating control actuation of the expansion valve to mitigate the disturbance effect. Both simulation and experimental validations of the proposed control strategy were carried out with a 3-ton variable-speed heat pump. Test results have shown that the proposed controller is effective in improving the superheat regulation performance for three load balancing scenarios, i.e., compressor speed step changes, frequency regulation and photovoltaic smoothing. The demonstrated performance gains include fast post-disturbance recovery of the superheat with the settling time shortened by 63% to 83%, complete avoidance of wet compression and enhanced transient energy efficiency.

Performance Analysis of a Cloud Computing System using Queuing Model with Correlated Task Reneging

Queuing theory has been extensively used in the modelling and performance analysis of cloud computing systems. The phenomenon of the task (or request) reneging, that is, the dropping of requests from the request queue often occur in cloud computing systems, and it is important to consider it when developing performance evaluations models for cloud computing infrastructures. Majority of studies in the performance evaluation of cloud computing data centres with the use of queuing theory do not consider the fact that the tasks could be removed from queue without being serviced. The removal of tasks from the queue could be due to the user impatience, execution deadline expiration, security reasons, or as an active queue management strategy. The reneging could be correlated in nature, that is, if a request is dropped (or reneged) at any time epoch, and then there is a probability that a request may or may not be dropped at the next time epoch. This kind of dropping (or reneging) of requests is referred to as correlated request reneging. In this paper we have modelled a cloud computing infrastructure with correlated request reneging using queuing theory. An M/M/1/N queuing model with correlated reneging has been used to study the performance analysis of the load balancing server of a cloud computing system. The steady-state as well as the transient performance analyses have been carried out. Important measures of performance like average queue size, average delay, probability of task blocking, and the probability of no waiting in the queue are studied. Finally, some comparisons are performed which describe the effect of correlated task reneging over simple exponential reneging.

Toward Optimal Partial Parallelization for Service Function Chaining

The emergence of Network Function Virtualization (NFV) and Service Function Chaining (SFC) together enable flexible and agile network management and traffic engineering. Due to the sequential execution nature of SFC, the latency would grow linearly with the number of functions. To resolve this issue, <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">function parallelization</i> has recently been proposed to enable independent functions to work simultaneously. Existing solutions, however, assume all the function instances are installed in the same physical machine and, thus, can be parallelized with only a little overhead. Nowadays, most of the networks deploy function instances in distributed servers for load balancing, parallelization across different servers would, in fact, introduce a non-negligible cost of duplicating or merging packets. Hence, in this work, we propose PPC ( <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Partial Parallel Chaining</i> ), which only parallelizes functions if parallelization can indeed reduce the latency after considering function placement and the required additional parallelization cost. To this end, we design two schemes, <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">partial parallelism enumeration</i> and <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">instance assignment</i> to identify the optimal partial parallelism that minimizes the latency. Our simulation results show that PPC effectively adapts the degree of parallelism and, hence, outperforms both sequential chaining and full parallelism in any general scenario. Overall, the latency reduction can be up to 47.2% and 35.2%, respectively, as compared to sequential chaining and full parallelism.

Edge Server Placement for Service Offloading in Internet of Things

With the rapid development of the Internet of Things, a large number of smart devices are being connected to the Internet while the data generated by these devices have put unprecedented pressure on existing network bandwidth and service operations. Edge computing, as a new paradigm, places servers at the edge of the network, effectively relieving bandwidth pressure and reducing delay caused by long-distance transmission. However, considering the high cost of deploying edge servers, as well as the waste of resources caused by the placement of idle servers or the degradation of service quality caused by resource conflicts, the placement strategy of edge servers has become a research hot spot. To solve this problem, an edge server placement method orienting service offloading in IoT called EPMOSO is proposed. In this method, Genetic Algorithm and Particle Swarm Optimization are combined to obtain a set of edge server placements strategies, and Simple Additive Weighting Method is utilized to determine the most balanced edge server placement, which is measured by minimum delay and energy consumption while achieving the load balance of edge servers. Multiple experiments are carried out, and results show that EPMOSO fulfills the multiobjective optimization with an acceptable convergence speed.

Security-Aware computation offloading for Mobile edge computing-Enabled smart city

Smart city has obtained increasing attention from both academic and industry which has the potential to improve human living standards. A smart city consists of a great number of smart devices which are generating large amounts of data and emerging applications all the time. However, the computing capacity of smart devices are limited. Fortunately, the emergence of MEC can solve the above problem. However, the resources of edge servers in MEC are limited and the capabilities of edge servers are heterogeneous. It is important to improve the average resource utilization of all edge servers and keep load balancing of edge server cluster simultaneously. On the other hand, quite a few numbers of applications are delay-sensitive, it is necessary to ensure the security of these applications. In this paper, we consider joint optimization of mobile device and edge server in MEC-enabled smart city, improving the overall performance of the system. Technically, a new multi-objective computation offloading method is implemented to reduce time consumption, energy consumption, and keep load balancing of edge servers, as well as increase average resource utilization of edge servers while meeting the deadline constraint of delay-sensitive applications. Sufficient experiments have been conducted to prove the effectiveness and superiority of our proposed method in different scenarios.

Elastic Performance Test Method of Web Server in Cloud Computing Environment

In the process of traditional web server elastic performance test, the performance of tracking data is poor, which leads to excessive server jitter and poor accuracy of elastic test results. Therefore, this paper studies the network server elastic performance test method in cloud computing environment. In this method, cloud monitoring technology is used to track the operation data of Web server. According to the multi platform call mode of Web server, a statistical regression model is established to determine the elasticity measurement index. The load balancing algorithm is used to test the server load balancing to obtain the elasticity value of Web server. Experimental results: in the whole running period of Web server, the jitter times of the proposed method are 15.6066 times, 16.5600 times and 16.5733 times lower than those of the three traditional methods, respectively. It can be seen that the new test method can accurately track the response ability of the server and obtain more accurate elasticity value.

Load Balancing Server and Homomorphic Encryption in Internet of Things

User demand for Internet of Things (IoT) services is ever increasing. The growing user demand can lead to an escalation of server workloads and faces the threat of theft of critical data. Consequently, a system is necessary to balance the server load and is protected with data encryption. In this study, we designed a system to share server workloads using load balancing methods. The load balancing technique uses open-source web server software. The system is equipped with data security using a homomorphic encryption algorithm from AES on the sender's side. The system embeds in an IoT telemedicine apparatus. During testing, we analyze the error requests that arrive at each server for the HTTP GET and POST methods. We also evaluate the speed of data encryption and decryption. The results showed that server load balancing reduces the number of error requests for the GET method by 97%. Meanwhile, the number of error requests for the POST method decreases by 66.75%. Observations reveal that the average homomorphic encryption speed, computation time, and decryption time are 15.66 ms, 764.18 µs, and 362.49 µs, respectively.

Application of DASH client optimization and artificial intelligence in the management and operation of big data tourism hotels

The hotel uses modern information technology to establish a customer data group, which contains information about the customer's information on space configuration, color design, etc., and uses scientific operation methods to accelerate the rapid, transformation and upgrading of the hotel. The rapid development of information technology represented by the big intelligent mobile cloud has a lot of functions, can meet the needs of different customers and diversified characteristics at the same time, and provides a complete, stable and accurate information integration platform for hotel system management. The main research directions of DASH include client adaptive algorithm design, multi-client fairness and resource utilization issues, client improvement under multiple servers, and server load balancing. The development and application of scientific and technological information has broken through the development path of the hotel management system, providing new development space and opportunities for the improvement of the hotel management system, but at the same time it will also face more new challenges. Therefore, this article focuses on the characteristics of the tourism accommodation management system under the big data environment, and analyzes the development of the hotel management system under the network environment. The purpose is to provide some reference and reference significance.

Traffic-aware adaptive server load balancing for software defined networks

Servers in data center networks handle heterogenous bulk loads. Load balancing, therefore, plays an important role in optimizing network bandwidth and minimizing response time. A complete knowledge of the current network status is needed to provide a stable load in the network. The process of network status catalog in a traditional network needs additional processing which increases complexity, whereas, in software defined networking, the control plane monitors the overall working of the network continuously. Hence it is decided to propose an efficient load balancing algorithm that adapts SDN. This paper proposes an efficient algorithm TA-ASLB-traffic-aware adaptive server load balancing to balance the flows to the servers in a data center network. It works based on two parameters, residual bandwidth, and server capacity. It detects the elephant flows and forwards them towards the optimal server where it can be processed quickly. It has been tested with the Mininet simulator and gave considerably better results compared to the existing server load balancing algorithms in the floodlight controller. After experimentation and analysis, it is understood that the method provides comparatively better results than the existing load balancing algorithms.