Min-Max Normalization

description17 papers

group29 followers

lightbulbAbout this topic

Min-Max Normalization is a data preprocessing technique used to scale numerical features to a specific range, typically [0, 1]. It transforms each feature by subtracting the minimum value and dividing by the range (maximum - minimum), ensuring that all features contribute equally to the analysis and improving the performance of machine learning algorithms.

lightbulbAbout this topic

Key research themes

1. How does Min-Max normalization impact clustering-based delineation of management zones in precision agriculture?

This research area investigates the role of Min-Max normalization and other normalization methods in improving the clustering of agricultural spatial data, which is essential for delineating management zones (MZs). Since clustering algorithms like fuzzy C-means rely on similarity measures sensitive to variable scales, normalization can critically affect the quality of MZ delineation, influencing economic and environmental outcomes.

Normalization of data for delineating management zones

by Nelson Betzek

2022, Computers and Electronics in Agriculture

Key finding: The study experimentally demonstrates that when multiple crop-related variables with different scales are clustered using Euclidean distance in fuzzy C-means, normalization is necessary. Among tested methods, Min-Max... Read more

articleView Paper downloadDownload

COMPARATIVE ANALYSIS OF NORMALIZATION TECHNIQUES IN THE CONTEXT OF MCDM PROBLEMS

by Ahmet Aytekin

2021, Decision Making: Applications in Management and Engineering

Key finding: Analyzing normalization impact on Multicriteria Decision Making problems, the paper discusses that Min-Max normalization is robust for transforming heterogeneous criteria scales into a dimensionless, comparable range. The... Read more

articleView Paper downloadDownload

Performance Comparison of New Adjusted Min-Max with Decimal Scaling and Statistical Column Normalization Methods for Artificial Neural Network Classification

by Saichon Sinsomboonthong

2023, International Journal of Mathematics and Mathematical Sciences

Key finding: Through empirical evaluation across six diverse datasets using artificial neural network classifiers, new adjusted Min-Max normalization variants were benchmarked against classical Min-Max, decimal scaling, and statistical... Read more

articleView Paper downloadDownload

keyboard_arrow_downShow more

2. What are the comparative effects of normalization methods on neural network performance, particularly focusing on Min-Max normalization and its variants?

This area explores how different data normalization techniques affect the training and predictive performance of neural networks, especially backpropagation networks. Since normalization impacts learning convergence and accuracy, detailed comparisons among Min-Max, Z-score, median-MAD, decimal scaling, and adjusted Min-Max provide actionable insights on method selection for optimizing neural network-based models.

A Normalization Methods for Backpropagation: A Comparative Study

by Adel Eesa

2021, Science Journal of University of Zakho

Key finding: The study applies multiple normalization methods, including Min-Max, Norm, Decimal scaling, Mean-Man, Median-Mad, and Z-score normalization to real UCI datasets for backpropagation neural networks. Results show Mean-Mad and... Read more

articleView Paper downloadDownload

A Normalization Methods for Back propagation: A Comparative Study.pdf

by Adnan Mohsin Abdulazeez

2019, Article

Key finding: Repeating and expanding on the 50679832 study, this paper confirms that Min-Max normalization is a simple yet effective method in transforming input/output vectors for feedforward neural networks trained by backpropagation,... Read more

articleView Paper downloadDownload

Performance Comparison of New Adjusted Min-Max with Decimal Scaling and Statistical Column Normalization Methods for Artificial Neural Network Classification

by Saichon Sinsomboonthong

2023, International Journal of Mathematics and Mathematical Sciences

Key finding: Extending previous comparative analyses, this paper experimentally shows that adjusted Min-Max normalization variants can outperform classical Min-Max in terms of accuracy and mean square error in neural network... Read more

articleView Paper downloadDownload

Sequential Normalization: Embracing Smaller Sample Sizes for Normalization

by Neofytos Dimitriou

2023, Information

Key finding: In neural network optimization contexts, advanced normalization layers inspired by Min-Max concepts like BatchNorm and GhostNorm normalize input feature distributions within mini-batches. GhostNorm and the proposed Sequential... Read more

articleView Paper downloadDownload

keyboard_arrow_downShow more

3. How can normalization be mathematically formulated and regularized for advanced data modeling problems including dimensionality reduction and eigenvalue computation?

This research theme deals with normalization formalization and its integration into algorithmic solutions for complex mathematical problems such as multidimensional scaling (MDS) and eigenvalue problem solving. The studies emphasize regularization techniques, interval data normalization, and normalized system transformations to improve computational stability and solution accuracy.

REGULARIZED MULTIDIMENSIONAL SCALING WITH RADIAL BASIS FUNCTIONS

by Sohana Jahan

2019, Journal of Industrial and Management Optimization

Key finding: The paper tackles normalization within nonlinear multidimensional scaling (MDS) using radial basis functions (RBF), highlighting the need to select RBF centers effectively. By treating center selection as a multitask learning... Read more

articleView Paper downloadDownload

On the Normalization of Interval Data

by Graçaliz Dimuro

2022, Mathematics

Key finding: This study reveals that straightforward application of interval arithmetic for normalization can inflate uncertainty ranges. It proposes axioms that any interval normalization arithmetic should satisfy to avoid worsening data... Read more

articleView Paper downloadDownload

Regularized Normalization Methods for Solving Linear and Nonlinear Eigenvalue Problems

by Chih-Wen Chang

2024, Mathematics

Key finding: The paper introduces normalization as a mechanism to convert homogeneous eigenvalue problems into nonhomogeneous systems, facilitating eigenvalue and eigenvector computation via regularized normalization coupled with... Read more

articleView Paper downloadDownload

Relaxation Subgradient Algorithms with Machine Learning Procedures

by Lev Kazakovtsev

2023, Mathematics

Key finding: By embedding machine learning-inspired regularization into relaxation subgradient minimization methods, this work uses normalization concepts as part of iterative learning algorithms to solve high-dimensional nonsmooth convex... Read more

articleView Paper downloadDownload

keyboard_arrow_downShow more

All papers in Min-Max Normalization

An Intelligent Dynamic Time Efficient Throttled Load Balancing Algorithm for Cloud Computing Environments

by Arthur Ndlovu

2024

descriptionView Paper arrow_downwardDownload

Pattern Discovery for Multiple Data Sources Based on Item Rank

by Dr. Arti Deshpande

2024, International Journal of Data Mining & Knowledge Management Process

Retail company's data may be geographically spread in different locations due to huge amount of data and rapid growth in transactions. But for decision making, knowledge workers need integrated data of all sites. Therefore the main... more

descriptionView Paper arrow_downwardDownload

Optimized Extreme Gradient Boosting with Remora Algorithm for Congestion Prediction in Transport Layer

by MECS Press and

2024, International Journal of Computer Network and Information Security (IJCNIS)

Transmission control protocol (TCP) is the most common protocol found in recent networks to maintain reliable communication. The most popular transport protocol in use today is TCP that cannot fully utilize the ability of the network... more

descriptionView Paper arrow_downwardDownload

The High Performance of a Task Scheduling Algorithm Using Reference Queues for Cloud- Computing Data Centers

by Chin-ling Chen

2024, Electronics

As the network technology continues to grow at a high rate of speed, the traditional network topology is improved with novel distributed topologies such as the Cloud computing network. A cloud computing environment consists of a huge... more

descriptionView Paper arrow_downwardDownload

Cluster Energy Optimization: An Algorithmic Approach

by Kuntal Mukherjee

2024, International Journal of Computer Applications

In fact, Gartner projected global revenue for cloud computing to reach almost $150 billion by 2014. However, The 2011 market is already approx $68 billion globally. With increase in web technologies and Internet, a proportional increase... more

descriptionView Paper arrow_downwardDownload

QOS Differentiation of Various Cloud Computing Load Balancing Techniques

by Navdeep Singh

2023, Communications on Applied Electronics

With an increase in the demands the Cloud computing has become one of the ongoing scalable approach to fulfill the cloud based neccesties. The biggest advantage of the cloud computing is the ability to overcome the infrastructural... more

descriptionView Paper arrow_downwardDownload

Improved Hybrid Algorithm Approach based Load Balancing Technique in Cloud Computing

by Dr. Srinivasa Rao Gundu

2023, Global journal of computer science and technology

The routine life of modern citizens is completely dominated by the computer aided services. The computer aided services depends on information and communication technologies. The success behind this cloud computing are data centers with... more

descriptionView Paper arrow_downwardDownload

Simulation of Load Balancing Algorithms using Cloud Analyst

by Azizkhan F Pathan

2023, International journal of engineering research and technology

Computing chips away at the premise of sharing of assets to accomplish intelligence. Instead of saving data over the proprietary hard drive, cloud computing facilitates the customers to store the data on remote database. Customer can... more

Figure 1: Basic structure of weighted round robin algorithm It operates on the principle of the round robin method, as the name suggests. After receiving a request from a client, the data center controller regulator directs the round robin load balancer to allocate another virtual machine (VM) for the purpose of processing the request. The weighted round robin algorithm is depicted in Fig. 1 as a basic representation of its structure.

Figure 2: Active Monitoring Load Balancing regarding the virtual machine. Subsequent to getting the reaction, the data center controller refreshes the assignment table by lessening the portion number | for that VM. In the wake of getting the reaction, the data center controller refreshes the allocation table by lessening the distribution number | for that VM.

This algorithm makes use of a throttle load balancer (TLB) to keep track of the amount of load and heap on each virtual machine in turn. In this case, each virtual machine is assigned to complete only one task at a time. Other task can be assigned only if the present work has been done successfully. The various steps associated with throttle algorithm are as per the following: The TLB maintains the index table which consists of information about all the VMs and their state. State of VM represents whether VM is available or busy. The client requests data center controller to allocate the appropriate VM to perform particular job. The data controller then demands TLB to allocate the VM.

happened to send an application or a site on the cloud and needed to test the administrations and burden that your item can deal with and furthermore tune its presentation to beat bottlenecks prior to gambling with organization, then, at that point, such assessments could be performed by basically coding a recreation of that climate with the assistance of different adaptable and versatile classes given by the CloudSim bundle, liberated from cost. Both exploration-based cloud improvement and industry-based cloud advancement have profited from the CloudSim venture's far and wide application. When taken from a calculated standpoint, CloudSim offers class definitions that address data centers, physical hosts, virtual machines, cloud administrations, cloud clients, inward centers networks, and the energy consumption of physical hosts, data centers components, and virtual machines from one perspective and virtual machines from the other. However, CloudSim allows for the dynamic insertion of simulation elements as well as the provision of message-passing applications and data centers network topologies on the other hand [17] [18]. The basic architecture of CloudSim is shown in fig. 4. Figure 4: Architecture of CloudSim

Figure 5: Architecture of Cloud Analyst Cloud Analyst is a reproduction apparatus planned in light of CloudSim. It gives a graphical UI that will uphold geologically appropriated huge scope cloud applications. The general reason for Cloud expert is to concentrate on the conduct of such applications under various organization arrangements. One of the fundamental goals of Cloud Analyst is to isolate the reproduction trial and error practice from a programming exercise. Cloud Analyst is based upon CloudSim toolbox by expanding its functionalities. Notwithstanding the CloudSim parts, the central parts are a Graphical User Interface (GUI) pack, as displayed in fig. 5.

In virtual machine migration, the main goal is to distribute a load of an overloaded data center among different data centers. It is just like the process of unloading the heavy machine by shifting its load to various different machines. We consider a whole machine to be a file or series of files using virtualization, move virtual computers across real machines thanks to this feature.

Figure 9: DC Hourly Average Processing Times Figure 8: UB Hourly Average Response Time

Table 1, 2, and 3 depict how we configured the cloud analyst tool for the Application Deployment, Data Center configurations, and User Base using cloud analyst tool. The world is divided into six distinct regions or districts in cloud analyst. In configuration of simulator we used simulation time setting for 60.0 minutes and Closest Data Center policy as “Service Broker Policy”.

Table 2: Application Deployment Configuration Table 3: Data Center Configuration

Spread Current Execution Algorithm (ESCE), and throttle respectively. Based on the simulation results, we discovered that the RR and ESCE algorithm have nearly identical overall response times in practice. While in the case of Throttled policy it creates a large difference between it and the other two Round Robin Algo:

Throttle Algo Table 5: Overall Response time of ESCE

Table 8: Overall DC Request Serving Time of ECSE

descriptionView Paper arrow_downwardDownload

Privacy Preserving in Data Mining by Normalization

by Lord Sir. Prince Khan

2023, International Journal of Computer Applications

Extracting previously unknown patterns from massive volume of data is the main objective of any data mining algorithm. In current days there is a tremendous expansion in data collection due to the development in the field of information... more

descriptionView Paper arrow_downwardDownload

Task-Scheduling Algorithms in Cloud Environment

by sekhar behera

2023, Advances in Intelligent Systems and Computing

The cloud is a technique based on the expansion of the possibilities and capabilities of the user's device by providing resources to implement the user's tasks. So, there has to be a mechanism to organize and distribute tasks for... more

Figure 1. Scheduling algorithm set task associated set resource (STASR) ne — af — “f The distribution of resources and tasks in groups helps to identify cost value where the us ‘an adjust within the scope of community resources and tasks, and in the case of user reques vere repeatedly within a certain range of tasks and resources groups can this wave of this us equests had to be automatically and Assume to those groups where are the exception reques rom the user of the queue and because it has been designated in advance by default ar utomatically. In the case, that group of resource groups are busy or any other circumstanc vhich increased the demand for them and became the tasks in a long wait, and to solve suc roblems we can assign group from resources groups as a virtual or resource on the level ach group are tasks by directing,, and here we can raise the efficiency of the algorithm and th eads to the lifting of valuable performance computerized network cloud. leads to the lifting of valuable performance computerized network cloud.

Figure 1. (b). The Relationship between execution time and Completion time in algorithm (STASR) Figure |. (a). The Relationship between The strategy that algorithm works on (STASR) is the proportionality between the execution time, and completion time, as in Figure 3 (a, b), If the execution time increases, the completion time decreases. It helps tasks and resources to distribute in the groups, in addition, the compatibility between tasks, and resource capacity to handle the requirements levels. As for the request that is presented from the task's user is frequently appointed to the resource groups, in the case if the resource was busy, the request is transferred to the backup resources, these backup resources are only used in case the resource’s groups are busy, also if the several tasks in waiting list, so, the tasks are transferred to the backup resources groups. in waiting list, so, the tasks are transferred to the backup resources groups.

Figure 4. Distribute the tasks to resources at the In the figure 4, represents the distribute the tasks to resources, based on that every resource represents a group, then ignoring the other resources in the same group, the tasks distribution to resources on this basis represents, in the figure 4, The value of Makespan is 21, while in the figure 5 the tasks distribution to resources based on the completion time, algorithm (RASA), the value of Makespan is 30.

‘igure 5. Distribute the tasks to resources at tl completion time, the algorithm (RASA) Figure 5. Distribute the tasks to resources at the

Figure 9. Distribution of tasks on individual resources, algorithm (STASR) Table 6.The Distribution Tasks to the Resources. Algorithm (Rasa)

Figure 7.Tasks distribution to the resources, Algorithm (RASA), in groups. Figure 6. The Tasks Distribution to the Resources, Algorithm (STASR), In Groups.

Figure 8.Distribution of tasks on individual resources

Figure 10. The Possibility of Distribution between the Tasks and Resources

Table 1. The Comparison between Algorithms in the Cloud Computing.

Table 2.The Relationship between the Execution Time, And Completion Time in (STASR) Table 3. Properties, Specifications and Distribution of Resources in Group

Table 4. Characteristics, Requirements of the Tasks, And Distributed In Groups.

As in the table (5) , tasks are distributed on the resources, based on the completion time, and commensurate with the execution time, as shown in Figure (1), which, shows the mechanism of the algorithm's work where the resources and tasks belong to groups. The resource is determined based on the compatibility between the resource group and tasks, also according to the requirements of the task.

Table 5.Distribute the Tasks to the Resources at the Completion Time The First Scenario

Table 7. Distribution of Tasks on Individual Resources, Algorithm (STASR)

descriptionView Paper arrow_downwardDownload

Review of Hybrid Load Balancing Algorithms in Cloud Computing Environment

by Doris Chinedu Asogwa

2022, Cornell University - arXiv

In cloud computing environment, load balancing is a key issue which is required to distribute the dynamic workload over multiple machines to make certain that no single machine is overloaded. In recent research, many organizations lose... more

descriptionView Paper arrow_downwardDownload

Pattern Discovery for Multiple Data Sources Based on Item Rank

by anjali mahajan

2022, International Journal of Data Mining & Knowledge Management Process

descriptionView Paper arrow_downwardDownload

The High Performance of a Task Scheduling Algorithm Using Reference Queues for Cloud- Computing Data Centers

by Chin-Ling Chen

2022, Electronics

Figure 1. The infrastructure of cloud computing. Therefore, cloud computing is the most important trend currently. The service providers can supply large computing resources, development platforms, and services, as shown in Figure 1. In various fields, the type of cloud computing can be sorted into three categories according to its applications:

Figure 2. The example of hierarchical virtualization topology.

Figure 4. Reference Queue based Cloud Service Architecture (RQCSA).

Figure 5. The Fitness Service Queue Selection Mechanism (FSQSM) flow chart.

there are three new coming tasks are sent to the CSC2. Therefore, by checking the RQ, the two RQ with the same remaining amount will be assigned for the tasks based on the rotation defined by the algorithm. As a result, the task which requires a five-unit time to be finished will be sent to the CM). In Round 4, the CSC2 already finished the prior task, and the CI distributes a new coming task to the CMB. In this case, because the CSC1, CSC2, and CSC3 are all in idle, the CM;, CM), and CM3 will distribute tasks to the CSC1, CSC2, and CSC3, respectively.

Figure 7. Queue waiting time, Cluster Manager (CM) = 16, queue size = 1024. The first discussion will be on the Random and Round-Robin algorithms. When the task numbers gradually increase starting from 1000, the queue waiting time of those two algorithms is much longer than the FSQSM. The main reason for the long waiting time is that those two algorithms do not consider the remaining amount of CM queue, which frequent task reassignments and reduce the performance of the entire system. Moreover, the tasks are distributed by randomly selecting the CMs

PRR EN RE Ey, CREE NAAEEE DA, eae Niee MMAR RR EE CERNE MEE Ge However, when testing the makespan, the experiment is performed based on the number of tasks between 2000 to 10,000 in order to obtain a more accurate analysis of the pros and cons of the three algorithms, as shown in Figure 8. The Random algorithm has a longer makespan because its random selecting method may select an RQ with no remaining amount when distributing tasks, in which the tasks may be missing and lead to a task re-distribution. Under the Round-Robin method, although it selects an RQ by rotation, it still does not take the RQ’s current remaining amount into account. Therefore, the makespan of the Round-Robin is still longer than the FSQSM. The proposed FSQSM considers the remaining amount of all RQs and selects the RQ with the most amount to distribute tasks which makes the makespan much shorter. The goal of testing the number of tasks is to verify the number of executed/completed services of the proposed FSQSM, Random and Round-Robin in a specific period. As the simulating time increases, the Random and Round-Robin serves fewer tasks than the FSQSM, and the reason for this is that the first two algorithms do not take the current remaining amount of RQ into account when distributing tasks and the system will be slowed down because the selected RQ may be fully loaded and causes task re-distributions. On the other hand, the FSQSM takes the remaining amount of RQ to avoid the drawback. There are some inconsistent results that occur in Figure 9, and the reason for this is that the completion time of each task is randomly defined. That means that if the FSQSM happens to serve tasks with shorter completion times, it will be able to handle more tasks, otherwise, it handles fewer tasks.

Figure 9. Number of tasks, CM = 16, queue size = 1024. remaining queue size of the CM; under each of the three algorithms will be calculated in low, medium, and high task-amount levels, respectively. The parameters and calculation of the MAD are shown in Table 3 and Equation (1). Subsequently, the results of the experiment are shown in Figures 10-19 and are used to calculate the MAD values of each of the three algorithms. With the MAD values, the distributions of the three algorithms can be verified.

Table 3. The parameters of the Mean Average Deviation (MAD). Figure 10. Remainder queue size of CM; for 1000 tasks.

Figure 11. Remainder queue size of CM; for 2000 tasks.

Figure 13. Remainder queue size of CM; for 4000 tasks. Figure 12. Remainder queue size of CM; for 3000 tasks.

Figure 14. Remainder queue size of CM; for 5000 tasks.

Figure 15. Remainder queue size of CM; for 6000 tasks.

Figure 17. Remainder queue size of CM; for 8000 tasks. Figure 16. Remainder queue size of CM; for 7000 tasks.

Figure 18. Remainder queue size of CM; for 9000 tasks.

Figure 20. Mean Average Deviation (MAD) for all tasks

Figure 19. Remainder queue size of CM; for 10,000 tasks.

Table 1. The FSQSM parameters and definitions. In Figure 6, the following four rounds have demonstrated the task execution and scheduling process caused by new coming tasks. The first two rounds describe the distribution mechanism of the FSQSM and show the example of RQ searching when the CM is fully loaded. The third and fourth rounds illustrate how to select the RQ by rotation when the RQs have the same remaining amount. Moreover, this example not only assumes that the execution time of each task is randomly generated and marked on the task but also defines the unit time as the CSC execution time in each round. In Round 1, the CM3 queue is fully loaded. The CSC1 and CSC3 are still executing the previous tasks and needs one-unit time. Now, a new coming task enters a fully loaded CM3, so the RQ mechanism is started and searches for available nodes. In the case of Round 1, the task will be assigned to CSC) and perform services

Table 2. The specifications of the experimental host and analysis tools. In the experimental environment, there will be 16 CMs to manage all the service nodes in the cloud. And the queue size of each CM will be 1024 slots. Next, three different task amounts will be assigned: Low (<3000 tasks), Medium (4000-7000 tasks), and High (>8000 tasks), and the experiment will perform 1000 simulations on each amount level, respectively. In the trail environment, all execution time is randomly assigned and lost tasks will be distributed again. waiting time, and makespan, will be executed. Moreover, this experiment even uses the Mean Average Deviation (MAD) to calculate the distributed task numbers and verify if the FSQSM distributes tasks to the CM evenly to avoid the overload of a single CM in which the overall efficiency will be lower [24-26]. The experimental host and analysis tools are listed in Table 2.

descriptionView Paper arrow_downwardDownload

Privacy Preserving in Data Mining by Normalization

by Prince Khan

2022, International Journal of Computer Applications

descriptionView Paper arrow_downwardDownload

Improved Hybrid Algorithm Approach based Load Balancing Technique in Cloud Computing

by Dr. Srinivasa Rao Gundu

2022, Global journal of computer science and technology

descriptionView Paper arrow_downwardDownload

Review of Hybrid Load Balancing Algorithms in Cloud Computing Environment

by Chiamaka Ijeoma Chukwuneke

2022

Figure 1:Load Balancing Technique (Karthika, Kanakambal & Balasubramaniam, 2015) Load balancing is a computer networking method to distribute workload across multiple computers or a computer cluster, network links, central processing units, disk drives, or other resources, to achieve optimal resource utilization, maximize throughput, minimize response time, and avoid overload (Kaur &Bansal,2013) Load balancing appears to be the major challenge in cloud computing, due to heterogeneous nature of cloud environment where resource pool is on increase (Gabi, Ismail & Zainal, 2015). The load balancer accepts multiple requests from the client and distributing each of them across multiple computers or network devices based on how busy the computer or network device is. Load balancing helps to prevent a server or network device from getting overwhelmed with requests and helps to distribute the work (Gabi, Ismail & Zainal (2015). Providing an efficient load balancing in cloud computing enables efficient resource utilization, achieved higher user satisfaction and also prioritizes users by applying appropriate scheduling criteria (Katyal & Mishra,2012).

descriptionView Paper arrow_downwardDownload

Implement of Dynamic Time Quantum Shortest Load First Scheduling for Efficient Load Balancing

by Enakshmi Nandi

2022, International Journal of Cloud-Computing and Super-Computing

Cloud Computing is a modern eminent technology in this era. Cloud service providers requires handling vast request for scaling up the size of cloud, but Cloud Computing faces critical challenges on its glorious future. Those challenges create problem for exploring for its perfect realization. Load Balancing is one of the hindrances in that case. Load Balancing is a process of distributing workload among different others nodes of distributed system to improve the perfect utilization of resources and job response time. Load Balancing should be done properly, as failure in any one of the node can lead of unavailability of data. There are already different research works has been done on Load Balancing using various scheduling techniques such as Round Robin (RR), First-Come-First-Served (FCFS), Shortest Load First (SLF) etc for reduce the execution time, reduce average waiting time for giving better quality of services in Cloud Computing. In this paper new scheduling has been approached ,that is Dynamic Shortest Load First Algorithm (DSLF).Shortest Load first is one of the sophisticated scheduling algorithm, for reduce the average waiting time compare to other existing algorithm and give better load balancing but this has some drawbacks. The burst time of processes have to predict before CPU start the execution. But it is impossible to predict the burst time of the process before the process complete the execution. If the process did not end their execution in definite predicted burst time then which algorithm will execute, then this question will come. In case of FCFS, starvation will occur, though the average waiting time will increase in that case. In my approach the main concept is, if the process remain in ready queue after ending of their predicted Burst Time, a new quantum will create dynamically and each unfinished process will get the amount of time quantum to execute in CPU if any process does not completed after execution of that amount of time (Time Quantum) it will go to tail of the ready queue and waiting for its turn [1].

descriptionView Paper arrow_downwardDownload

Privacy Preserving in Data Mining by Normalization

by Prince Khan and

2022, International Journal of Computer Applications

descriptionView Paper arrow_downwardDownload

Review of Hybrid Load Balancing Algorithms in Cloud Computing Environment

by Moses Okechukwu Onyesolu

2022

descriptionView Paper arrow_downwardDownload

Load balancing with preemptive and non-preemptive task scheduling in cloud computing

by Shahrulniza Musa

2022, 2017 IEEE 3rd International Conference on Engineering Technologies and Social Sciences (ICETSS)

In Cloud computing environment the resources are managed dynamically based on the need and demand for resources for a particular task. With a lot of challenges to be addressed our concern is Load balancing where load balancing is done for... more

a grouped into clusters. In terms of security, the most secure cloud service provider is the one in which the presence of physical server is not known to the virtual instance that is running. If this is the case, then a determined hacker faces a tough time which is targeting to hack the particular task present in the physical environment. The task needs to have a Service Level Agreement [4] with the third parties to be more secure. The information when is to be accessed by a wide variety of users by using IoT [5] in different domains like in health care etc., it should be readily available. rig. 1 Cioud COMpullng service WLodels. The above figure shows the various cloud computing service models and the one responsible for managing them. The ability to understand and to fully trust the availability, reliability, and performance of the cloud are the key reasons for the organizations to show their interest to move to the cloud. For all the service models, still, the aim of each service model is to serve the client in an interrupted way by providing the best services in an efficient manner. Load balancing is the primary concern that needs the load to be balanced by dynamically allocating the resources to each required node and ensures to maintain the QoS.

Fig. 5 Data center loading at each data center using RR policy. Using the Round Robin policy, the response time at each user base can be seen in fig.3, the processing times at each data center can be seen in fig.4 and the loading on each of the data center per hour can be seen in fig.5. Because the policy used is Round Robin which is preemptive, gives a fair share to each task without making any task to wait for a long time in the queue. However, the problem with this policy is that even a small task has to wait for its turn if it is the last one to be processed.

Fig. 9 Data center loading at each data center using ESCE policy.

Fig. 10 Response time by region using throttled policy. Fig. 11 Response time at each userbase using throttled policy.

Fig. 13 Data center loading at each data center using throttled policy.

Fig. 15 Data Center processing times of all policies The response times of all the policies are graphically represented in fig. 14, which shows that equally spread current execution policy has an average less response time but the max response time is more than others.

In throttled scheduling, some of the tasks are killed at the data center, and they are migrated to other for processing without the knowledge of the user. The results in fig. 10 show the response times at each user base by region, and it is observed that the response time is more here as some of the tasks are migrated to other data center. Fig.11 shows the response time at each user base graphically. The processing time taken by each data center can be observed in fig.12, where it is more than the other policies. In fig.13 the data center loading is shown which is also more as the tasks are throttled based on the load at each data center and migrated to other data centers for execution. Fig. 14 Response times of all policies.

The resources in a cloud computing environment can be CPUs, memory storage, networks, services, and applications. The scheduling policies manage the resources. There are various scheduling policies categorized as preemptive and non-preemptive scheduling. The resources needed to complete the task are estimated, and they are checked for availability using the above formula. An efficient scheduling algorithm is the one who finds the resources always available with no other task waiting in the queue for its completion. Cloud Analyst, is a simulator tool used here to analyze the results. Some of the scheduling policies like Round Robin(RR), Equally spread current execution load(ESCE), throttled are used for simulation. Various loads are distributed on the data centers in different regions, with user base assigned to each data center. The metrics used are response time, data center servicing times and data center loading. From fig 2, it is clear that for each userbase a data center is allocated and so the average response times are almost close to each other.

With equally spread current execution load policy, the user bases are assigned to each data center equally with no data center overloaded with tasks at the very beginning. So there is no preemption here. As there is no preemption, the results in fig. 6, show that the average response time at each user base in each region is less than the response time in other policies. Fig.7 graphically represents the response time in milliseconds. Fig.8 shows the processing time is also less than the service times in other policies. Data center loading is seen in Fig.9 serving the number of requests per hour.

Fig. 12 Service times at each data center using throttled policy.

descriptionView Paper arrow_downwardDownload

A Survey on Load Balancing Algorithms in Cloud Computing

by Priyanka Sarma

2022, International Journal of Computer Sciences and Engineering

Cloud computing is sharing various computing resources rather than having local servers or personal devices to handle applications over internet. There is lots of data stored on cloud and multiple requests come for resources, due to this... more

Fig 1: Cloud Computing System anywhere, anytime on the basis of “only pay for what you use”. (Software is provided as a Service) are provided online using Internet Connection. These services are provided

IIIT COMPARISON TABLE OF EXISTING LOAD BALANCING ALGORITHMS IV CONCLUSION

descriptionView Paper arrow_downwardDownload

Load balancing with preemptive and non-preemptive task scheduling in cloud computing

by Muhammad Shadab Alam

2022, 2017 IEEE 3rd International Conference on Engineering Technologies and Social Sciences (ICETSS)

descriptionView Paper arrow_downwardDownload

Privacy Preserving in Data Mining by Normalization

by Tarique Ahmad

2022

descriptionView Paper arrow_downwardDownload

Privacy Preserving in Data Mining by Normalization

by shameemul haque

2022

descriptionView Paper arrow_downwardDownload

A new scheduling algorithm for server farms load balancing

by Shafigh Parsazad

2022, International Conference on Industrial and Information Systems

This paper describes a new scheduling algorithm to distribute jobs in server farm systems. The proposed algorithm overcomes the starvation caused by SRPT (Shortest Remaining Processing Time). This algorithm is used in process scheduling... more

descriptionView Paper arrow_downwardDownload

Improve Performance by Using Load Balancing Algorithm to Reduce Response Time and Processing Time on Cloud Computing

by Kalpesh Wandra

2022

Cloud computing is one of the incredible technology which enable the new vision for IT industry. Nowadays, it has become a strong alternative for startup large as well as small scale organizations that only use the resources which... more

descriptionView Paper arrow_downwardDownload

Load balancing with preemptive and non-preemptive task scheduling in cloud computing

by Dr. Mohammad Riyaz Belgaum

2022, 2017 IEEE 3rd International Conference on Engineering Technologies and Social Sciences (ICETSS)

descriptionView Paper arrow_downwardDownload

R Esearches Tailoring and Instantiation of Mde Process Models : A y Model-Based Approach

by Alassane BAH

2022

The adoption of MDE in software development has drastically changed the way of developing software products. It gives a way of automating software development while reducing production time. While software companies need continually to... more

descriptionView Paper arrow_downwardDownload

Scheduling and Load Balancing Techniques in Cloud Computing: A Survey

by Mithun Dsouza

2022

Cloud computing has become popular due to its attractive features. The load on the cloud is increasing tremendously with the development of new applications. Load balancing is an important part of cloud computing environment which ensures... more

connected to the Internet. These devices submit their request and receive the response without any delay [3]. Figure 1, it shows different devices (mobiles, PCs, laptops) connect and access the data from a cloud at any given time.

Figure 2. Load balancing in cloud computing

descriptionView Paper arrow_downwardDownload

A Review on Dynamic Resource Allocation Based on Lease Types in Cloud Environment

by Navtej Ghumman

2021

Load balancing is one of the essential factors to enhance the working performance of the cloud service provider. Cloud Computing is an emerging computing paradigm. It aims to share data, calculations, and service transparently over a... more

descriptionView Paper arrow_downwardDownload

AMO Based Load Balancing Approach in Cloud Computing

by madhukar dubey

2021, IOSR Journal of Computer Engineering

Cloud Computing is an evolving paradigm with altering definitions, but for this research task, it's defined as a virtual infrastructure that presents shared data and communication era services. Load Balancing is any other essential issue... more

descriptionView Paper arrow_downwardDownload

AMO Based Load Balancing Approach in Cloud Computing

by Madhukar Dubey

2021, IOSR Journal of Computer Engineering

This work implemented over cloudsim Table.3 implemented over cloudsim

Fig.3 Enter number of VMs for starting cloudsim Result fable 4 Comparison Result between Existing | and Proposed Techniques — =.= oe | = Amotest [Java Application] C:\Program Files\Java\jre7\bin\javaw.exe (Mar 12 2017 3:43:43 PM) Starting CloudSim simulation Example using Animal Migration optimisation...

Fig.4 Comparision of base and proposed technique

DOI: 10.9790/066 1-1902046266 loud architectures are underlying on infrastructure that is exploited only when its required that draw the needed resources on-demand and perform a specific job, then abandon the unneeded resources and usually influence of them later the job is completed. SaaS is a multi-tenant platform that utilizes general resources and a sole example of together the object code of an application along with the underlying database to support many customers simultaneously. The prime global players of SaaS are SalesForce, IBM, Microsoft, and Oracle. PaaS give developer with platform comprising each the systems and atmosphere containing the end-to-end life cycle of testing, deploying hosting and developing of complex web applications. Iaas is the delivery of computer IaaS. The usage depends on payment element is a key advantage of PaaS. Flexiscale, Rackspace and GoGrid etc are to name some instance for PaaS.

Table. 2 comparison of load balancing algorithms [4] IV. Animal Migration

descriptionView Paper arrow_downwardDownload

Comparing load balancing algorithms for web application in cloud environment

by Zakaria Benlalia

2021, Indonesian Journal of Electrical Engineering and Computer Science

Cloud computing has emerged as a new paradigm for providing on-demand computing resources and outsourcing software and hardware infrastructures. Load balancing is one of the major concerns in cloud computing environment means how to... more

descriptionView Paper arrow_downwardDownload

QOS Differentiation of Various Cloud Computing Load Balancing Techniques

by mohit birdi

2021, Communications on Applied Electronics

descriptionView Paper arrow_downwardDownload

QOS Differentiation of Various Cloud Computing Load Balancing Techniques

by mohit birdi

2021, Communications on Applied Electronics

descriptionView Paper arrow_downwardDownload

A Comparative Study of Various Load Balancing Techniques in Cloud Computing

by reena panwar

2021

In recent years, Cloud Computing has been an emerging computing model in the IT industry. Cloud computing opens up the world of computing to a broader range of uses and increases the ease of use by giving access to anything and anywhere... more

In the PaaS models, cloud providers deliver a computing platform, typically including operating system, programming language execution environment, database, and web server. Fig1. Cloud Computing Services

descriptionView Paper arrow_downwardDownload

Pattern Discovery for Multiple Data Sources Based on Item Rank

by Arti Deshpande

2021, International Journal of Data Mining & Knowledge Management Process

descriptionView Paper arrow_downwardDownload

A Novel Pattern Merger Algorithm for generating Actionable Rules for Multi-Source Combined Mining

by Arti Deshpande

2021, International Journal of Computer Applications

The idea of combined mining is very useful and flexible for identifying in-depth patterns through combining sets of items from multiple datasets or using multiple data mining techniques. The identified combined patterns disclose in-depth... more

descriptionView Paper arrow_downwardDownload

Analytical Approach to Estimate and Optimize the Defects in Prepreg Lamination for a Component Produced by Stacking of Blanks from a Cutter

by IJRASET Publication

2021, International Journal for Research in Applied Science & Engineering Technology

Carbon Fiber-reinforced polymer composite materials are fast gaining ground as preferred materials for the aero engine components. For the easiness of manufacturing instead of traditional fiber and resin layup process, the composites are... more

descriptionView Paper arrow_downwardDownload

Improved Hybrid Algorithm Approach based Load Balancing Technique in Cloud Computing

by Dr. Srinivasa Rao Gundu

2020, Global Journal of Computer Science and Technology: B Cloud and Distributed

descriptionView Paper arrow_downwardDownload

Improved Implementation of Hybrid Approach in Cloud Environment

by Dr. Srinivasa Rao Gundu

2020, International Journal of Computer Sciences and Engineering

Cloud computing is a technological advancement which is based on the concept of dynamic resources provisioning using internet. Web 2.0 technologies play a central role. Hybrid approach to load balancing will provides better results in... more

descriptionView Paper arrow_downwardDownload

Predictive Software Model for Quality Assessment

by IJRASET Publication

2020, International Journal for Research in Applied Science and Engineering Technology IJRASET

Tender usually refers to the process whereby governments invite offers for plenty of projects that must be submitted within a finite deadline. Tendering is the process of providing an offer, bid or proposal, or expressing interest in... more

descriptionView Paper arrow_downwardDownload

Pleuroscopia

by Dr. PETRU EMIL MUNTEAN

2020, Viata Medicala

UUOMIPUCaAta Riscul de mortalitate este de 0,09%, cu o rata a complicatiilor commenti’ ocean ae hg it pene rar resceonees erbsas Somoe Principalele indicatii se refe- ra la: diagnosticarea revarsatelor pleurale de tip idiopatic, stadiali- zarea cancerului pulmonar, pleuro- deza, biopsia pulmonara si pericar- dica, stadializarea mezoteliomului. Principalele —_ contraindicatii sunt legate de capacitatea si com- petenta medicului de a efectua corect aceasta procedura. Urma- Vizualizarea cavitatii pleurale se realizeaza prin intermediul unui instrument optic numit pleuro- scop. Exist doua categorii diferite de pleuroscoape: rigide si semiri- gide. Alegerea instrumentului de- pinde de indicatia procedurii, insa majoritatea interventiilor pot fi efectuate cu un pleuroscop semiri- gid. Pleuroscopul semirigid este similar unui video-bronhoscop. Se compune dintr-un maner si un tub atasat, masoara 27 cm lungime si 7 mm in diametru. Primii 22 cm din tubul pleuroscopului sunt ri- gizi, iar cei 5 cm de la capatul distal bibbiograie Petru Emil Muntean, Petre Botianu, Mirela Liana Gliga, Dan Ulmeanu. Referent stiintific Joana Soare. Punctia pleurala. Bucuresti: Editura Etna, 2017. ISBN 978- 973-1985-26-8 rentru a presau mterventia, medicul evacueaza pana la 500 ml de lichid din spatiul pleural, prin toracenteza, si poate chiar induce un pneumotorax inainte de a in- sera un trocar. Este posibila intro- ducerea trocarului de la inceput, printr-o incizie la nivel intercostal, pentru a permite aspirarea lichi- dului pleural dupa introducerea trocarului. In situatia in care se suspecteaza o afectare maligna, se efectueaza o singura incizie la ni- vel tegumentar, in al cincilea pana la cel de-al saptelea spatiu inter- costal, de-a lungul peretelui lateral al hemitoracelui afectat. Lichidul pleural este evacuat, esantioane sunt trimise la laborator pentru

Fig. Cong Amiloidoza. A. Hematoxilina-cozina; B. Rosu de Congo; C. Birefringenta galben-verzuie (Coloratie Rosu de ‘0 in Itunina polarizata) nediteraneeana familiala), bolilor ntestinale cronice (boala Crohn), i unor neoplazii. Este consecinta lepunerilor in spatiile extracelula- ¢ aproteinei denumita amiloidul A eric (SAA), wn important reactant le faza acuta secretat de ficat prin ctiunea stimulatorie a citokinelor dentierea prezentei amiloidului precizarea naturii SAA a acestuia prin teste imunohistochimice spe- cifice. Evolutia AA depinde de cea a bolii inflamatorii subiacente. Tra- tamentele actuale - colchicina in fe- bra mediteraneeana familiala, aza- tioprina, corticoterapia, precum ce (Simona Stancu Hildegard, Spi- talul de Nefrologie ,Carol Davila’ Bucuresti), cardiologice (Ruxandra Jurcut, Institutul C.C. Iliescu), gas- troenterologice (Speranta Iacob. Institutul Clinic Fundeni), neuro- logice (Mirela Draghici, Institutul Clinic Fundeni), aspectele imagis

descriptionView Paper arrow_downwardDownload

A STUDY ON CLASSIFICATIONS OF CLOUD

by INTERNATIONAL JOURNAL OF ENGINEERING TECHNOLOGIES AND MANAGEMENT RESEARCH I J E T M R JOURNAL

2020, International Journal of Engineering Technologies and Management Research

"The interesting thing about cloud computing is that we've redefined cloud computing to include everything that we already do."-Larry Ellison Cloud computing is an emerging field that is expected to change the entire information... more

descriptionView Paper arrow_downwardDownload

Affinity and Association

by Tushar Gupta

2019

descriptionView Paper arrow_downwardDownload

IJCSE

by Dr. Srinivasa Rao Gundu

2019

Fig.1. Equally spread current execution algorithm working mechanism Equally spread current execution algorithm simulation result Fig.1. Equally spread current execution algorithm working mechanism

Fig. 2. RTEHVMLB Algorithm Working Mechanism be constructed using a combination of RTHEH Algorithm. In order to avoid the delay in response time and for the heterogeneous or different types hardware configurations of virtual machines it can be combined the RTH with equally spread current execution algorithm and call it as RTEH algorithm. It is observed that this algorithm also reduces the context Switch and reduces the turnaround time and increases the throughput.

Flowchart.1. RTEAHVMLB Algorithm Flowchart

Fig.3. RTEAH algorithm working mechanism element. For algorithm simulation: Six user bases are there: they are UB1,UB2,UB3,UB4,UB5,UB6 can be increased ' or decreased ,Four data centres are there: They are D1,D2,D3,D4 can be increased or decreased, Six regions are there: RO,R1,R2,R3,R4,R5,R6(Regions are fixed . Each data center is constructed with 15 Physical machines each have the following configuration. 8086 architecture, Linux operating system, Xen virtual machine manager, 204800 RAM(MB), 100TB Storage space,10,0000 Bandwidth, 4 Processors, 1000 MIPS Time shared virtual machines Scheduling Policy. Each physical machine is consolidated with 10 virtual machines. Each virtual machine is configured as 1000 MB Image size, 512 MB Memory, 1000 MB Bandwidth.

Fig.4. Load balancing in RTEAH Algorithm in comparison with other algorithms Graph

Table 1. Equally spread current execution algorithm simulation result Equally spread current execution algorithm simulation result

International Journal of Computer Sciences and Engineering RTEAH Algorithm in comparison with other algorithms Graph

descriptionView Paper arrow_downwardDownload

EDEMA PARU ppt

by Tita Garrien

2019

descriptionView Paper arrow_downwardDownload

Load Balancing Strategy using Round Robin Algorithm

by Hermanda Ihut Tua Simamora

2019

Current cloud computing environment serves in practically every field of our life. Be that as it may, while satisfying parts and loads of client solicitations it confronts couple of confinements to be overcome. Alongside giving us offices... more

ine. The CPU scheduler will then choose the following procedure in the readied line.

[Fig. 2] Example for RR Algorithm We assume five processes arriving at time=0, with increasing burst time (P1=4, P2=5, P3=2, The Gantt chart is shown below:

[Fig. 4] Example for Modified RR Step 8: repeat step 6 till all processes will not complete their execution.

[Table 1] Comparison Table of RP and MBR For that here it is represents a slight change in the round robin algorithm, called as nodified round robin algorithm. This modified round robin algorithm is the combination of shortest job first(SJF) and round robin(RR). By calculating the modified round robin the oarameters like high waiting time, response time are overcome. The modified round robin sives the better performance than round robin algorithm. The parameters like waiting time, srATniew: Gaaveem« a: « aawwac and bebel sassesslkes wal amandiack owsbiahwnan aes weasel

descriptionView Paper arrow_downwardDownload

Comparative Analysis of Load Balanceing-19.docx

by Dr. Muhammad Qasim

2019, ISSOS Publisher

Today cloud computing has become the cornerstone of information technology, the next stage in web development. It provides users with a great deal of computing and storage services over the Internet. The load balancing is necessary for... more

descriptionView Paper arrow_downwardDownload

Improve Performance by Using Load Balancing Algorithm to Reduce Response Time and Processing Time on Cloud Computing

by IOSR Journals

2019, IOSR JOURNALS