Multi-Objective Prioritized Task Scheduler Using Improved Asynchronous Advantage Actor Critic (a3c) Algorithm in Multi Cloud Environment

IEEE Access, Journal Year: 2024, Volume and Issue: 12, P. 11354 - 11377

Published: Jan. 1, 2024

Task scheduling is a crucial challenge in cloud computing paradigm as variety of tasks with different runtime processing capacities generated from various heterogeneous devices are coming up to application console which effects system performance terms makespan, resource utilization, cost. Therefore, traditional algorithms may not adapt this efficiently. Many existing authors developed task schedulers by using metaheuristic approaches solve problem(TSP) get near optimal solutions but still TSP highly dynamic challenging scenario it NP hard problem. To tackle challenge, paper introduces multi objective prioritized scheduler improved asynchronous advantage actor critic(a3c) algorithm uses priorities based on length tasks, and VMs electricity unit cost environment. Scheduling process carried out two stages. In the first stage, all incoming VM calculated at manager level second Priorities fed (MOPTSA3C) generate decisions map effectively onto considering schedule cost, makespan available Extensive simulations conducted Cloudsim toolkit giving input trace fabricated data distributions real time worklogs HPC2N, NASA datasets scheduler. For evaluating efficacy proposed MOPTSA3C, compared against techniques i.e. DQN, A2C, MOABCQ. From results, evident that MOPTSA3C outperforms for reliability.

Language: Английский

---

Techniques for load balancing throughout the cloud: a comprehensive literature analysis

International Journal of Computational and Experimental Science and Engineering, Journal Year: 2025, Volume and Issue: 11(1)

Published: Jan. 12, 2025

Recently, "Cloud-Computing (CC)" has become increasingly common because it's a new paradigm for handling massive challenges in versatile and efficient way. CC is form of decentralized computation that uses an online network to facilitate the sharing various computational computing resources among large number consumers, most commonly referred as "Cloud-Users (CUs)”. The burdens on "Cloud-Server (CS)" could be either light or too heavy, depending how quickly volume CUs their demands are growing. Higher response times high resource usage two many issues resulting from these conditions. To address enhance CS efficiency, "Load-Balancing (LB)" approaches very effective. goal LB approach identify over-loading under-loading CSs distribute workload accordingly. Publications have employed numerous techniques broad effectiveness solutions, boost confidence end CUs, ensure effective governance suitable CS. A successful technique distributes tasks within network, thereby increasing performance maximizing utilization. Experts shown abundance engagement this issue offered several remedies over past decade. primary extensive review article examine different variables provide critical analysis current techniques. Additionally, outlines requirements explores associated with context CC. Conventional insufficient they ignore operational efficiency “Fault-Tolerance (FT)” measures. present article, bridge gaps existing research, assist academics gaining more knowledge about

Language: Английский

---

Task-Driven Virtual Machine Optimization Placement Model and Algorithm

Future Internet, Journal Year: 2025, Volume and Issue: 17(2), P. 73 - 73

Published: Feb. 7, 2025

In cloud data centers, determining how to balance the interests of user and service provider is a challenging issue. this study, task-loading-oriented virtual machine (VM) optimization placement model algorithm proposed integrating consideration both VM user’s computing requirements. First, modeled as multi-objective problem minimize makespan loading tasks, rental costs, energy consumption centers; then, an improved chaos-elite NSGA-III (CE-NSGAIII) presented by casting logistic mapping-based population initialization (LMPI) elite-guided in NSGA-III; finally, CE-NSGAIII employed solve aforementioned model, further, through combination above sub-algorithms, CE-NSGAIII-based method developed. The experiment results show that Pareto solution set obtained using exhibits better convergence diversity than those compared algorithms yields optimized scheme with shorter makespan, less lower consumption.

Language: Английский

---

Improvement of the Approximate Method for Determining the Average Vertical Stress Increase Below the Rectangular Foundation Using Differential Evolution Algorithm

Arabian Journal for Science and Engineering, Journal Year: 2025, Volume and Issue: unknown

Published: April 23, 2025

Abstract External loads transferred from the structure's foundations to soil induce stress increases in stratum. Since within mass vary with depth and across plane at a given depth, approaches that estimate average increase under can be advantageous for effective foundation design. This study aims develop optimization-based approximate methods calculating vertical higher accuracy than conventional 2V:1H method rectangular different L/B ratios. For this purpose, projection 120 depths 12 ratios were numerically calculated using Boussinesq’s expressions. The model parameters of proposed models, such as expansion slopes (k or k 1 , 2 ) normalized critical (z cr /B), each ratio optimized differential evolution algorithm. three-parameter achieved highest accuracy, reducing RMSE values by an 53% compared method, while one-parameter reduced 9%. maximum absolute errors remained between 0.0217 0.0283, R greater 0.9972. Building upon improving presents practical novel provides more reliable accurate estimation flexible foundations, significantly errors. contributes geotechnical engineering prediction potentially leading economical safer designs.

Language: Английский

---

A Comprehensive Survey of MapReduce Models for Processing Big Data

Big Data and Cognitive Computing, Journal Year: 2025, Volume and Issue: 9(4), P. 77 - 77

Published: March 27, 2025

With the rapid increase in amount of big data, traditional software tools are facing complexity tackling which is a huge concern research industry. In addition, management and processing data have become more difficult, thus increasing security threats. Various fields encountered issues fully making use these large-scale with supported decision-making. Data mining methods been tremendously improved to identify patterns for sorting larger set data. MapReduce models provide greater advantages in-depth evaluation can be compatible various applications. This survey analyses map-reducing utilized processing, techniques harnessed reviewed literature, challenges. Furthermore, this reviews major advancements diverse types map-reduce models, namely Hadoop, Hive, Pig, MongoDB, Spark, Cassandra. Besides reliable approaches, also examined metrics computing performance among More specifically, review summarizes background its terminologies, types, different techniques, applications advance framework processing. study provides good insights conducting experiments field managing

Language: Английский

---

Efficient virtual machine placement in cloud computing environment using BSO-ANN based hybrid technique

Alexandria Engineering Journal, Journal Year: 2024, Volume and Issue: 110, P. 145 - 152

Published: Oct. 8, 2024

Language: Английский

---

An Energy-Efficient VM Selection Using Updated Dragonfly Algorithm in Cloud Computing

International Journal of Computer Theory and Engineering, Journal Year: 2024, Volume and Issue: 16(3), P. 76 - 86

Published: Jan. 1, 2024

Language: Английский

---

Power Management of Cloud Datacenter in Infrastructure Level via Efficinet Virtual Machine Placement by Utilizing Hybrid Genetic Algorithm

Published: July 13, 2023

The virtualization technology enables cloud datacenters to co-host multiple virtual machines (VMs) on a single physical machine (PM) meet the need of different users. Efficient VM placement (VMP) schemes can significantly decrease residual power consumption in infrastructure level. This paper formulates VMP an integer linear optimization programming problem with management perspective. To solve this NP-Hard problem, hybrid genetic algorithm (HGA) is presented. verify effectiveness proposed HGA, it was tested against other state-of-the-arts scenarios. results shows that HGA beats approaches terms and reduction total significantly.

Language: Английский

---

Cloud Datacenter Selection Using Service Broker Policies: A Survey

Computer Modeling in Engineering & Sciences, Journal Year: 2023, Volume and Issue: 139(1), P. 1 - 41

Published: Dec. 29, 2023

Amid the landscape of Cloud Computing (CC), Datacenter (DC) stands as a conglomerate physical servers, whose performance can be hindered by bottlenecks within realm proliferating CC services. A linchpin in CC’s performance, Service Broker (CSB), orchestrates DC selection. Failure to adroitly route user requests with suitable DCs transforms CSB into bottleneck, endangering service quality. To tackle this, deploying an efficient policy becomes imperative, optimizing selection meet stringent Quality-of-Service (QoS) demands. Amidst numerous policies, their implementation grapples challenges like costs and availability. This article undertakes holistic review diverse concurrently surveying predicaments confronted current policies. The foremost objective is pinpoint research gaps remedies invigorate future development. Additionally, it extensively clarifies various methodologies employed CC, enriching practitioners researchers alike. Employing synthetic analysis, systematically assesses compares myriad techniques. These analytical insights equip decision-makers pragmatic framework discern apt technique for needs. In summation, this discourse resoundingly underscores paramount importance adept policies selection, highlighting imperative role performance. By emphasizing significance these modeling implications, contributes both general its practical applications domain.

Language: Английский

---

Enhanced Linear Regression Models for Resource Usage Prediction in Dynamic Cloud Environments

International Journal of Advanced Computer Science and Applications, Journal Year: 2024, Volume and Issue: 15(1)

Published: Jan. 1, 2024

In response to the diverse resource utilization patterns observed across enterprises, this study proposes of adaptable cloud services. A novel system framework is presented, capturing and logging consumption at discrete intervals. Subsequently, recorded data serves as input for a linear regression model, functioning machine learning tool predict in forthcoming intervals, leveraging historical stored within module. To bolster resilience various effective meta-heuristic techniques are integrated alongside conventional methodology, facilitating more accurate anticipation overloaded or under-loaded conditions before their occurrence real-world scenarios. Simulations demonstrate that hybrid algorithm, named Whale Optimization Algorithm-based Linear Regression (WOA-LR), outperforms Genetic Algorithm-Linear (GA-LR), Particle Swarm Optimization-Linear (PSO-LR), JAYA-LR, traditional (LR) achieving desired objective functions significantly reducing Mean Squared Error (MSE). This approach holds promise prediction optimization dynamic environments.

Language: Английский

---