Load Balancing Algorithm in Cloud Computing
Cloud computing is the one of the technology which helps to use the resources from different sources especially like data storage,using hardware and software sources. It incurs fewer amounts for using these facilities. Many software companies are using this type of facilities because of low cost. They need to pay only the amount for using the facilities.
They don’t need to pay for maintenance. This attracts many business enterprises and all of them jumped towards using the cloud. So automatically load will be increased. The problem will occur. So in this paper we will review some of the algorithm for load balancing in cloud computing and we will compare and find the best one which helps to overcome this problem.
Keywords: Load Balancing, RRA, AMLB, TLBA
The cloud computing is the internet based development that satisfies many organizations because of its cheaper cost. It is attracted by many users. Because it provides high speed, good security, high processing speeds, storage facility etc at reasonable cost. This was the main advantage of using cloud. The data stored is distributed in the network.
The data may be nearer or it may be stored in remote. The programs can be launched in the network and the result can be seen in our network browser. So many data centers are there to store the cloud data.
In cloud computing the load should be balanced effectively so that it will have maximum throughput with minimal response time. Otherwise the use of cloud computing won’t attract many organizations. To reduce this problem we will discuss some algorithms and reviews to perform the cloud computing effectively.
- Basic load balancing Algorithms in Networks
Load balancing means to distribute the load evenly. So the resources are fairly allocated with high level of user satisfaction. There are many reviews and they are using different algorithms to improve the load balancing in cloud computing. Higher resource utilization and proper load balancing will reduce the resource consumption. So there will be quicker response times.
[A] Round Robin Algorithm
It is the simplest algorithm et al . It uses the time interval. Each node is given a certain amount of time in that the node can perform their operation. So quantum of time interval plays an important role and its drawback is extra load to the scheduler.
Fig 1: RRA
[B] Active Load Balancing and Monitoring
This is the equally spread load balancing algorithm. The ALBM equally divides the loads to the VM. It maintains the index which contains the id of virtual machine, and amount of load allocated to it. When the request comes the index from top to least VM is found. Then it returns the VM id to ALBM. In this way the load is equally distributed. The main disadvantage is it always uses the least VM id.
Fig 2: AMLB
[C] Throttled Load Balancing Algorithm
In this algorithm the VM is allocation first with one task. After successful completion first task, then only the second task will be allocated to the VM machine. It also maintains an index of all VM. If any request arrives it searches from top to bottom in index table until it finds the VM. If it finds the task is allocated. Otherwise it will return the null value. The disadvantage is if it returns null then the data centers has to process the work.
Fig 3: TLBA
- Literature Survey
Cloud Analyst Simulator
In this the author Almothana Khodar et al  says by comparing the above three algorithms and finds the best algorithm which balances the load effectively. Above three algorithms are combined by using the cloud analyst tool. By changing some parameters it will give the better result. The terminology of emulator is as follows,
Regions: It is divided into six regions
User Base: It is used to create traffic
Data Processing Center: It accepts and process the request
VM load Balancer: It is responsible for the distributing loads to the data centers.
The author uses this tool to analyze the best performance of these tools by changing some parameters. Finally the author concludes that the TLBA is the best load balancing algorithm.
R K Banyal et al  says that the enhanced algorithm will reduce the load and it will evenly distribute the load. He used the tool kit called as cloudsim tool kit to perform effective load balancing. It has the framework it supports both operation of system and cloudsim components.
Nowadays most of the organization are using the cloud computing and they are paying the amount for what they have utilized. So it plays an important role in internet. Without internet the cloud computing is not possible. It was the new emerging technique to develop the business at peak.
The CloudSim architecture is follows,
Fig 4: CloudSim
The components of cloudsim are data center, data center broker, host, VM and cloudlet. It has a preprocessing capability, memory, storage, and policy for allocating processor to VM. The java is used here to implement CloudSim. Here it checks for available free processor in VM. If it finds it will allocate the task.
If there is no host free then it will check for the processor. If it finds then it will allocate to the processor with free host. The author proposed that the main aim is to check free data centers and available processing elements.
Shahbaz Afzal et al  tells that the load balancing can be viewed in two way. They are if the load is overloaded or it is under loaded. In this aspect the author has taken some sample databases and analyzed the algorithms in all metrics. Most of the algorithm is implemented in the simulator.
The load balancing algorithm should be focused on different dimensions in which QOS metrics and algorithm complexity should be considered. The author has stated and reviewed the problem of load balancing algorithms and he also suggests in what dimension the algorithm should be viewed.
Amrita Jyoti et al  has stated that the cloud computing are the very important sources for convenient resource for sharing applications, storage, servers, computing etc. There are three types which help to service.
They are SaaS, PaaS, IaaS. There are different types of cloud. They are public cloud, private cloud, community cloud, hybrid cloud. The author states that there are two types of load balancing technique. They are static and dynamic technique. In static the previous system knowledge is known like processing power, data, memory etc.
The best for this technique is round robin. He also suggested another algorithm called as shortest job algorithm. In this approach the shortest job is selected first. The waiting time of shortest job is very less. The next algorithm is Min-Min algorithm.
The time required to complete the task is calculated and minimum time job completion is selected first. Next is max min algorithm. It is same like min-min algorithm but with some difference. The next algorithm is combination of two algorithms the first one is opportunistic load balancing algorithm and another one is min-min algorithm.
It is combined to get better efficiency. The dynamic load balancing algorithm performs dynamically. In this the first algorithm is energy aware load balancing algorithm. The second one is modified active monitoring load balancer algorithm and another one is genetic algorithm.
The author also summarized about other algorithm like honey bee foraging technique, ant colony algorithm and other load balancing techniques like particle swarm optimization technique active monitoring algorithm.
In this paper the author finally concluded that different types of algorithms are summarized and to perform efficient and quick access, all algorithms should be used at different way. Mainly the load balancing plays an important role in cloud computing.
The author mallikarjuna et al  has proposed that the load balancing plays an important role in data storage and hardware usage. The author compares the dynamic and static load balancing techniques and suggests that dynamic load balancing technique provides the efficient use of clod computing.
Fig 5: Emerging Technologies in future
The author strictly says that dynamic load balancing algorithm and dynamic nature inspired load balancing algorithm is best suited for new emerging technologies like IOT, big data computing and self-learning systems. In future automatic dynamic load balancing algorithm is used and load balancing can be done efficiently and the resource utilization will be more for the entire task.
Sambit Kumar Mishra et al  states that the cloud computing plays an important role in internet technologies. It was the vast development in internet. Among that the load balancing problem is the important one that is when overloaded or under loaded.
The different types of algorithms techniques are used to balance the load correctly. Here in this paper the author suggest using heuristic technique to balance the load.
To balance the load so many thing should be considered. They are throughput, reliability, accuracy, predictability, make span, response time, associated cost, energy consumption, scalability, fault tolerance, migration time are the some of the attributes that should be considered in handling load balancing technique.
The author has stated different types of static and dynamic algorithm under homogenous and heterogeneous conditions. To apply heuristic based approach the cloud environment should be implemented in the cloudsim simulator.
The three algorithms are compared. They are round robin, throttled, active monitoring algorithm. Nowadays day by day internet usage is increased. In the same the use of cloud architecture is also increased.
They mainly uses data storage, processing, computing, executing the application and for other purposes.
The main problem in using cloud computing is the load balancing. Load balancing means equally sharing the task in the distributed environment. It is the biggest challenge in cloud computing. To achieve all this three algorithm are compared and the best one is used in an efficient way. The active monitoring algorithm is the best one to handle the load in an efficient way.
- Proposed Work
We have reviewed different types of static and dynamic algorithm. All algorithms are focused on the maximum throughput, better performance, minimum response time etc. The dynamic algorithm can be used effectively by changing some parameters in that we achieve best load balancing technique.
So load balancing plays an important role in new technology. The proposed model should be developed and implemented in the cloud analyst tool and we can measure the maximum throughput and response time.
- Conclusion and Future Work
I conclude here that load balancing is the important and key factor which plays in the cloud computing. In future the cloud computing plays an important role in growing environment.
Because the cloud environment provides all services in a cheaper amount. So many business environments are engaged in using cloud. The data that is stored is very secured and safe. The only thing is load balancing and it can be performed efficiently by using the algorithm. The metrics are also should be considered while generating the algorithm.
 Violetta N. Volkova ,Liudmila, V. Chernenkaya, Elena N. Desyatirikova, Moussa Hajali, Almothana Khodar, Alkaadi Osama, Load Balancing in Cloud Computing, IEEE Conference of Russian Young Researchersin Electrical and Electronic Engineering 2018.
 Ojasvee Kaneria, R K Banyal, Analysis and Improvement of Load Balancing in Cloud Computing, International Conference on ICT in Business Industry and Government, 2016.
Shahbaz Afzal and G. Kavitha, Load Balancing in Cloud Computing – A Hierarchical Taxonomical Classification, Journal of Cloud Computing : Advances, Systems and Applications 2019, Volume 8 Issue 22.
 Amrita Jyoti, Manish Shrimali, Shailesh Tiwari, Harivans Pratap Singh, Cloud Computing using Load Balancing and Service broker policy for IT service: A Taxonomy and Survey , Journal of Ambient Intelligence and Humanized Computing, 2019.
 B. Mallikarjuna, D. Arun Kumar Reddy, The Role of Load Balancing Algorithms in Next Generation of Cloud Computing , Journal of Advanced Research in Dynamical & Control Systems, 2019, Volume 11, Issue 07.
 Sambit Kumar Mishra , Bibhudatta Sahoo, Priti Paramita Parida, Load Balancing in Cloud Computing : A Big Picture, Journal of King Saud University – Computer and Information Sciences, 2018 Volume 32.
 Chen, S.L., Chen, Y.Y. and Kuo, S.H., 2017. CLB: A novel load balancing architecture and algorithm for cloud services. Computers & Electrical Engineering, 58, pp.154-160.
 Bhole, R., Singh, H.J., Khamkar, P., Joshi, P. and Bendbhar, R., 2017. Load Balancing in Cloud Computing Using Autonomous Agents. Imp. J. Interdiscip. Res, 3, pp.237-239.
 Kumar, M., Dubey, K. and Sharma, S.C., 2018. Elastic and flexible deadline constraint load balancing algorithm for cloud computing. Procedia Computer Science, 125, pp.717-724.
 Peng, H., Han, W., Yao, J. and Fu, C., 2018, October. The realization of load balancing algorithm in cloud computing. In Proceedings of the 2nd International Conference on Computer Science and Application Engineering (pp. 1-5).
 Dash, S., Panigrahi, A. and Sabat, N.R., 2019. Performance Analysis of Load Balancing Algorithm in Cloud Computing.
 Alam, M. and Khan, Z.A., 2017. Issues and challenges of load balancing algorithm in cloud computing environment. Indian J. Sci. Technol, 10(25), pp.1-12.
 Narayanan, S.S., Ramakrishnan, M. and Basha, M.S., 2017. Efficient load balancing algorithm for cloud computing using divisible load scheduling and weighted round Robin methods. Advances in Natural and Applied Sciences, 11(1), pp.13-19.
 Kumar, M. and Sharma, S.C., 2017. Dynamic load balancing algorithm for balancing the workload among virtual machine in cloud computing. Procedia computer science, 115, pp.322-329.
 Chien, N.K., Son, N.H. and Loc, H.D., 2016, January. Load balancing algorithm based on estimating finish time of services in cloud computing. In 2016 18th International Conference on Advanced Communication Technology (ICACT) (pp. 228-233). IEEE.
 Padmavathi, M. and Basha, S.M., 2017, June. Dynamic and elasticity ACO load balancing algorithm for cloud computing. In 2017 International Conference on Intelligent Computing and Control Systems (ICICCS) (pp. 77-81). IEEE.
 Hamdani, M., Aklouf, Y. and Bouarara, H.A., 2019, March. Improved fuzzy Load-Balancing Algorithm for Cloud Computing System. In Proceedings of the 9th International Conference on Information Systems and Technologies (pp. 1-4).
 Singh, S.I., Abraham, T.C. and Iyengar, N.C.S.N., 2016. A review: Different improvised min-min load balancing algorithm in cloud computing environment. Journal of Computer and Mathematical Sciences, 7(11), pp.540-550.
Academic Research Writing Arm of Global Research Services.