ABSTRACT With the increasing compute and data demands of intricate scientific applications, cloud computing has emerged as a key solution for executing scientific workflows while adhering to Quality of Service (QoS) parameters. Developing an efficient and cost‐effective solution, especially for large‐scale applications, continues to be an immensely challenging task. To handle this challenge, this research presents a scheduling algorithm, an opposition‐based chaotic enhanced African vulture workflow scheduler (OCEAVWS), to reduce the makespan and financial cost incurred in executing scientific workflows. The proposed approach initially employs opposition‐based learning (OBL) techniques to create a high‐quality initial population, utilizes a logistic chaotic map that assists in selecting phases in the African vulture optimization algorithm (AVOA), and is further enhanced by dimension learning hunting (DLH). The strength of the proposed method is assessed on the WorkflowSim tool on four different scientific workflows. The experimental results show substantial improvements, including an 18.03% reduction in makespan and an 8.07% decrease in financial costs. The evaluation metrics indicate that our approach achieves lower average relative deviation index (ARDI) values in all cases, lower S‐metric values in 83% of scenarios, and higher hypervolume in 92% of the evaluated scenarios. Statistical validation using Wilcoxon and Friedman tests confirms that OCEAVWS significantly outperforms five other state‐of‐the‐art multi‐objective optimization approaches, demonstrating its effectiveness in scientific workflow scheduling.

ABSTRACT Cloud computing has changed the technology landscape for over a decade and led to an astounding growth in the number of applications it may be used for. Consequently, there has been a significant spike in the demand for improved algorithms to schedule workflows efficiently. These were mostly concerned with heuristic, metaheuristic, and hybrid approaches to workflow scheduling that mostly suffer from the problem of local optima entrapment. Due to such heavy traffic on the cloud resources, there is still a need for less computationally complex approaches. In light of this, this article proposes a novel approach: a multi‐objective Modified Local Escaping Archimedes Optimization (MLEAO) algorithm for workflow scheduling. This strategy involves initialization of the population of Archimedes Optimization algorithm through the HEFT algorithm to provide an inclination towards the solutions with improved makespan while achieving a cost‐efficient workflow scheduling decision and avoiding the problem of local optima entrapment using a local escaping operation. To validate the efficacy of our approach, we conducted extensive experiments using scientific workflows as benchmarks. Through our investigations, we significantly improved makespan, cost, resource utilization, and energy consumption. Moreover, the effectiveness of our proposed approach is also verified by performance metrics such as hypervolume, s‐metric, and dominance relationships between the proposed and state‐of‐the‐art approaches.

Summary A difficult problem in the service‐oriented computing paradigm is improving task scheduler policy or resource provisioning.In order to increase the performance of cloud applications, this article primarily focuses on tasks for resource mapping policy optimization. With the aim of reducing makespan and execution overhead and increasing the average resource utilization, we suggested an efficient independent task scheduler employing supervised neural networks in this paper. The suggested ANN‐based scheduler uses the status of the cloud environment and incoming tasks as inputs to determine the optimal computing resource for a given assignment as a result that assembles our goal. We proposed a novel algorithm in this paper that uses a hybrid methodology based on a swarm intelligence algorithm (PSO) in combination with a machine learning technique (ANN). PSO is used to prepare the train and test dataset for the neural network. Results clearly state that suggested work achieves significant improvement to considered algorithms in makespan (45%–55%), average VM utilization (15%–20%), and execution overhead(20%–30%).

Abstract Cloud computing is a promising platform for executing scientific workflow applications. Commonly, the scientific workflow applications are complex in size and computation intensive. Task scheduling in clouds is a familiar NP‐complete problem. Efficient workflow application scheduling is critical for meeting multi‐objectives in cloud environments. By reason of its pivotal role, this problem has been widely studied, and many approaches have been developed. Most of the algorithms focus on schedule makespan and execution cost. The cloud data centers consume a large amount of energy while running workflow applications due to a lack of efficient scheduling algorithms to perform the task to a virtual machine (VM) mapping. Hence, there is a significant need to address energy utilization in the cloud data center (CDC) for two reasons such as data center operational cost optimization and to improve the environment. This article presents an energy and cost‐aware scheduling (ECWS) approach with the objectives of decreasing energy utilization, execution cost, and maximizing utilization of the resources. The ECWS algorithm is a heterogeneous earliest finish time (HEFT) based energy‐efficient heuristic for cloud scheduler. The ECWS algorithm includes three sub‐algorithms like RE calculation, RE Threshold selection, and slack algorithm. The effectiveness of the proposed algorithm is evaluated on the WorkflowSim tool using distinct workloads from various scientific areas. The experimental outcome exhibit that the proposed algorithm achieved significant energy conservation, maximized resource utilization, and cost‐saving when compared with related well‐known algorithms.

Abstract Cloud computing is a technology that is being used by scientists for the execution of huge scale workflows as it has various features such as availability, faster, cheaper, elasticity, and high accessibility. In addition, the users are charged on a pay per use basis which makes it more efficient to be used by various clients. Scheduling workflows in the cloud is a difficult task as it takes into account the dependencies of workflows along with quality of service (QoS) demands. The problem becomes more challenging when there are multiple QoS objectives which are often disagreeing to each other. A lot of research has been done to minimize makespan and execution cost under deadline and budget constraints. This article presents a workflow scheduling algorithm based on Jaya to minimize them both and returns solutions according to the weights a user gives. The algorithm has been compared with Min‐Min, Max‐Min, Round Robin, heterogeneous earliest finish time first, first come first serve, minimum completion time, dynamic heterogeneous earliest finish time first in WorkflowSim. The results are compared based on both cost and time. In the end, a trade‐off between the objectives has been shown. Results show that Jaya performs much better than other algorithms when averaged over several iterations.

With the unprecedented growth of Cloud Computing, data centers around the world have increased exponentially. The energy utilization in these data centers is becoming a concern, so the need for energy-efficient algorithms in the cloud has been on the top agenda for quite a while. Cloud providers use different energy management strategies to minimize energy utilization and to maximize ROI (Return On Investment) such as energy-efficient virtual machine (VM) placement. We have taken the problem for VM consolidation and worked on it for an energy-efficient algorithm. VM Consolidation uses live migrations of VMs during the execution of cloudlets so that underloaded physical servers can be switched off by transferring those VMs to other physical machines. VM consolidation in the cloud environment is a proven NP-hard problem. We have used Water Wave optimization (WWO) which is a meta-heuristic algorithm. Original WWO proposed for continuous space and so we have modified the parameters of the algorithm to apply for our problem of VM consolidation. Our approach produces a near-optimal solution using an objective function that minimizes energy consumption and increases the number of switched off servers.

Summary Cloud computing is an on‐demand service that can be accessed by a user according to his requirements through the Internet. Multiple users can request any amount of services, so scheduling of those services is a crucial task in cloud computing. Scheduling is a way of assigning the work to a computer resource. We have multiple tasks at a time that are waiting to be allotted to multiple computer resources. Various optimization algorithms have been used to do task scheduling so that total execution cost is minimized. In this paper, we have implemented Jaya optimization algorithm for workflow scheduling and have compared it with four nature‐inspired algorithms, namely, particle swarm optimization (PSO), genetic algorithm (GA), ant colony optimization (ACO), honey bee, and cat swarm optimization (CSO), keeping the fitness function same for all of them using CloudSim. Previously, work has been done on PSO, GA, ACO, honey bee, and CSO using different criteria. The results are compared on the basis of execution cost and makespan of the algorithm on both an independent set of tasks and a set of tasks that follow a workflow schedule. Benchmark functions such as Montage, CyberShake, Inspiral, and Sipht are used for workflow scheduling. It has been observed that Jaya outperforms the other algorithms as it produces similar results in the least amount of time as it converges very quickly.

Summary Maximization of availability and minimization of the makespan for transaction scheduling in an on‐demand computing system is an emerging problem. The existing approaches to find the exact solutions for this problem are limited. This paper proposes a task scheduling algorithm using ant colony optimization (MATS_ACO) to solve the mentioned problem. In this method, first, availability of the system is computed, and then, the transactions are scheduled using the foraging behavior of ants to find the optimal solutions. We also modify two known meta‐heuristic algorithms such as genetic algorithm (GA) and extremal optimization (EO) to obtain transaction scheduling algorithms for the purpose of comparison with our proposed algorithm. The compared results show that the proposed algorithm performs better than others.

Summary Balanced task allocation is one of the methods that can be used to maximize the performance and reliability in the on‐demand computing‐based transaction processing system. On‐demand computing is an increasingly popular enterprise model. It provides computing resources to the user as needed, which may be maintained within the user's enterprise, or made available by a service provider. The balanced task allocation in such environment is known to be an NP hard. The reliability is a measure of trustworthiness of the system while executing the task. So we derive the reliability formula for the on‐demand computing‐based transaction processing system considering resource availability. We propose the balanced task allocation based on social spider optimization (LBTA_SSO) method for this problem. The LBTA_SSO is based on the cooperative behavior of social‐spiders to find a collection of task allocation solutions. We modified five existing algorithms to obtain the task allocation algorithms; Honey Bee Optimization (HBO), Ant Colony Optimization (ACO), Hierarchical Load Balanced Algorithm (HLBA), Dynamic and Decentralized Load Balancing (DLB), and Randomized Algorithm respectively. Then, we compared the proposed algorithm with these modified algorithms. The results show that our algorithm works better than the modified existing algorithms.

Summary The scheduling of load balanced transactions is an emerging problem in on‐demand computing system. This problem can be studied considering several parameters such as resource availability, performance, and reliability. The existing approaches to find the exact solutions for this problem are limited. This paper presents a Honey Bee Optimization (HBO)–based method to solve the mentioned problem. In this method, first, the load of the system is balanced and then the transactions are scheduled using foraging behavior of honey bees to find the optimal solutions. We also modify four known scheduling algorithms such as Ant Colony Optimization (ACO), Hierarchical Load Balanced Algorithm (HLBA), Dynamic and Decentralized Load Balancing (DLB), and Randomized to obtain transaction scheduling algorithms for the purpose of comparison with our proposed algorithm. The compared results show that the proposed algorithm performs better than the modified existing algorithms.