Genetically-modified Multi-objective Particle Swarm Optimization approach for high-performance computing workflow scheduling
Abstract and Figures
Nowadays, scientific research, industry, and many other fields are greedy regarding computing resources. Therefore, Cloud Computing infrastructures are now attracting pervasive interest thanks to their excellent hallmarks such as scalability, high performance, reliability, and the pay-per-use strategy. The execution of these high-performant applications on such kind of computing environments in respect of optimizing many conflicting objectives brings us to a challenging issue commonly known as the multi-objective workflows scheduling on large scale distributed systems. Having this in mind, we outline in the present paper our proposed approach called Genetically-modified Multi-objective Particle Swarm Optimization (GMPSO) for scheduling application workflows on hybrid Clouds in the context of high-performance computing in an attempt to optimize Makespan and Cost. The GMPSO consists of incorporating genetic operations into the Multi-objective Particle Swarm Optimization to enhance the resulting solutions. To achieve this, we have designed a novel solution encoding that represents the task ordering, the task mapping and the resource provisioning processes of the workflow scheduling problem in hybrid Clouds. In addition, a set of particular adaptive evolutionary operators have been designed. Conducted simulations lead to significant results compared with a set of well-performed algorithms such NSGA-II, OMOPSO and SMPSO, especially, for the most-demanding workload of workflows.
Figures - uploaded by Hamza Gharsellaoui
Author content
All figure content in this area was uploaded by Hamza Gharsellaoui
Content may be subject to copyright.
... To optimize makespan, Cheikh et al. [30] proposed a hybrid heuristic scheduling algorithm by combining PSO and Extremal Optimization (EO), which used PSO to provide the initial solution for EO. PSOM [31] performed the mutation operator of GA on each particle to improve PSO. Nwogbaga et al. [32] applied PSO after GA for the population evolution. ...
... PSO with Mutation (PSOM) improves PSO by the mutation operator [31]. It performs the mutation operator on each particle after each evolution of PSO. ...
Task scheduling is still an open issue for improving the performance of cloud services. Focusing on addressing the issue, we first formulate the task-scheduling problem of heterogeneous cloud computing into a binary non-linear programming. There are two optimization objectives including the number of accepted tasks and the overall resource utilizations. To solve the problem in polynomial time complexity, we provide a hybrid heuristic algorithm by combing both benefits of genetic algorithm (GA) and particle swarm optimization (PSO), named PGSAO. Specifically, PGSAO integrates the evolution strategy of GA into PSO to overcome the shortcoming of easily trapping into local optimization of PSO, and applies the self-cognition and social cognition of PSO to ensure the exploitation power. Extensive simulated experiments are conducted for evaluating the performance of PGSAO, and the results show that PGSAO has 23.0–33.2% more accepted tasks and 27.9–43.7% higher resource utilization than eight other meta-heuristic and hybrid heuristic algorithms, on average.
... Some researchers worked on time parameters for workflow prioritization and allocation of resources. Hafsi et al. [57] have gone for a combined solution using a genetic algorithm and particle swarm optimization. The main advantage of this method is its high scalability in terms of demand and time-constrained problems, which is also considered in simulation. ...
Fog and cloud computing are emerging paradigms that enable distributed and scalable data processing and analysis. However, these paradigms also pose significant challenges for workflow scheduling and assigning related tasks or jobs to available resources. Resources in fog and cloud environments are heterogeneous, dynamic, and uncertain, requiring efficient scheduling algorithms to optimize costs and latency and to handle faults for better performance. This paper aims to comprehensively survey existing workflow scheduling techniques for fog and cloud environments and their essential challenges. We analyzed 82 related papers published recently in reputable journals. We propose a subjective taxonomy that categorizes the critical difficulties in existing work to achieve this goal. Then, we present a systematic overview of existing workflow scheduling techniques for fog and cloud environments, along with their benefits and drawbacks. We also analyze different workflow scheduling techniques for various criteria, such as performance, costs, reliability, scalability, and security. The outcomes reveal that 25% of the scheduling algorithms use heuristic-based mechanisms, and 75% use different Artificial Intelligence (AI) based and parametric modelling methods. Makespan is the most significant parameter addressed in most articles. This survey article highlights potentials and limitations that can pave the way for further processing or enhancing existing techniques for interested researchers.
... Hafsi, Gharsellaoui, and Bouamama (2022) proposed a genetically modified multiobjective particle swarm optimization method for scheduling high-performance computing workflows. Their method showed a significant 30% decrease in the time it takes to complete tasks and a 10% improvement in the use of resources [10]. ...
Cloud Computing (CC) offers abundant resources and diverse services for running a wide range of consumer applications, although it faces specific issues that need attention. Cloud customers aim to choose the most suitable resource that fulfills the requirements of consumers at a fair cost and within an acceptable timeframe; however, at times, they wind up paying more for a shorter duration. Many advanced algorithms focus on optimizing a single variable individually. Hence, an Optimized Resource Allocation in Cloud Computing (ORA-CC) Model is required to achieve equilibrium between opposing aims in Cloud Computing. The ORA-CC study aims to create a task processing structure with the decision-making ability to choose the best resource in real-time for handling diverse and complicated uses on Virtual Computers (VC). It will utilize a Modified Particle Swarm Optimization (MoPSO) method to meet a deadline set by the user. The fitness value is calculated by combining a base value with the enhanced estimation of resources based on the ORA-CC algorithm to create a robust arrangement. The ORA-CC technique's effectiveness is evaluated by contrasting it with a few current multi-objective restrictions applied to machine scheduling strategies utilizing the Cloudsim simulation. The comparison demonstrates that the suggested ORA-CC strategy offers more efficient resource allocation than other techniques.
... To prove the superiority of our method, we compare PGSAO with five classical and widely used meta-heuristic algorithms, GA [16], Differential Evolution (DE) [19], Artificial Bee Colony (ABC) [20], PSO [14] and Multi-Verse Optimizer (MVO) [21], and two hybrid meta-heuristic algorithms (GAPSO [22] and PSOGA) in solving ESDP. GAPSO is to perform GA in the first half of the evolutionary stage, and PSO in the second half. ...
There have been some works proposing meta-heuristic-based algorithms for the task offloading problem in Device-Edge-Cloud Collaborative Computing (DE3C) systems, due to their good performance than heuristic-based approaches. But these works don’t fully exploit the complementarity of multiple meta-heuristic algorithms. In this paper, we combine the benefits of both swarm intelligence and evolutionary algorithm, for designing a high-efficient task offloading strategy. To be specific, our proposed algorithm uses the iterative optimization framework of Particle Swarm Optimization (PSO) to exploit the cognitions of swarm intelligence, and applies the evolutionary strategy of Genetic Algorithm (GA) to preserve the diversity. Extensive experiment results show that our proposed algorithm has better acceptance ratio and resource utilization than nine of classical and up-to-date methods.KeywordsGenetic AlgorithmParticle Swarm OptimizationTask OffloadingEdge ComputingCloud Computing
Cloud computing provides solutions to a large number of organizations in terms of hosting systems and services. The services provided by cloud computing are broadly used for business and scientific applications. Business applications are task oriented applications and structured into business workflows. Whereas, scientific applications are data oriented and compute intensive applications and structured into scientific workflows. Scientific workflows are managed through scientific workflows management and scheduling systems. Recently, a significant amount of research is carried out on management and scheduling of scientific workflow applications. This study presents a comprehensive review on scientific workflows management and scheduling in cloud computing. It provides an overview of existing surveys on scientific workflows management systems. It presents a taxonomy of scientific workflow applications and characteristics. It shows the working of existing scientific workflows management and scheduling techniques including resource scheduling, fault-tolerant scheduling and energy efficient scheduling. It provides discussion on various performance evaluation parameters along with definition and equation. It also provides discussion on various performance evaluation platforms used for evaluation of scientific workflows management and scheduling strategies. It finds evaluation platforms used for the evaluation of scientific workflows techniques based on various performance evaluation parameters. It also finds various design goals for presenting new scientific workflow management techniques. Finally, it explores the open research issues that require attention and high importance.
In recent years, the development of new algorithms for multiobjective optimization has considerably grown. A large number of performance indicators has been introduced to measure the quality of Pareto front approximations produced by these algorithms. In this work, we propose a review of a total of 63 performance indicators partitioned into four groups according to their properties: cardinality, convergence, distribution and spread. Applications of these indicators are presented as well.
In recent algorithmic family simulates different biological processes observed in Nature in order to efficiently address complex optimization problems. In the last years the number of bio-inspired optimization approaches in literature has grown considerably, reaching unprecedented levels that dark the future prospects of this field of research. This paper addresses this problem by proposing two comprehensive, principle-based taxonomies that allow researchers to organize existing and future algorithmic developments into well-defined categories, considering two different criteria: the source of inspiration and the behavior of each algorithm. Using these taxonomies we review more than three hundred publications dealing with nature-inspired and bio-inspired algorithms, and proposals falling within each of these categories are examined, leading to a critical summary of design trends and similarities between them, and the identification of the most similar classical algorithm for each reviewed paper. From our analysis we conclude that a poor relationship is often found between the natural inspiration of an algorithm and its behavior. Furthermore, similarities in terms of behavior between different algorithms are greater than what is claimed in their public disclosure: specifically, we show that more than one-third of the reviewed bio-inspired solvers are versions of classical algorithms. Grounded on the conclusions of our critical analysis, we give several recommendations and points of improvement for better methodological practices in this active and growing research field.
In the last decades, global optimization problems are very common in many research fields of science and engineering and lots of evolutionary computation algorithms have been used to deal with such problems, such as differential evolution (DE) and particle swarm optimization (PSO). However, the algorithms performance rapidly decreases as the increasement of the problem dimension. In order to solve large-scale global optimization problems more efficiently, a lot of improved evolutionary computation algorithms, especially the improved DE or improved PSO algorithms have been proposed. In this paper, we want to analyze the differences and characteristics of various large-scale evolutionary optimization (LSEO) algorithms on some benchmark functions. We adopt the CEC2010 and the CEC2013 large-scale optimization benchmark functions to compare the performance of seven well-known LSEO algorithms. Then, we try to figure out which algorithms perform better on different types of benchmark functions based on simulation results. Finally, we give some potential future research directions of LSEO algorithms and make a conclusion.
Optimal scheduling of workflows in cloud computing environments is an essential element to maximize the utilization of Virtual Machines (VMs). In practice, scheduling of dependent tasks in a workflow requires distributing the tasks to the available VMs on the cloud. This paper introduces a discrete variation of the Distributed Grey Wolf Optimizer (DGWO) for scheduling dependent tasks to VMs. The scheduling process in DGWO is modeled as a minimization problem for two objectives: computation and data transmission costs. DGWO uses the largest order value (LOV) method to convert the continuous candidate solutions produced by DGWO to discrete candidate solutions. DGWO was experimentally tested and compared to well-known optimization-based scheduling algorithms (Particle Swarm Optimization (PSO), Grey Wolf Optimizer). The experimental results suggest that DGWO distributes tasks to VMs faster than the other tested algorithms. Besides, DGWO was compared to PSO and Binary PSO (BPSO) using WorkflowSim and scientific workflows of different sizes. The obtained simulation results suggest that DGWO provides the best makespan compared to the other algorithms.
Workflows are a commonly used model to describe applications consisting of computational tasks with data or control flow dependencies. They are used in domains of bioinformatics, astronomy, physics, etc., for data-driven scientific applications. Execution of data-intensive workflow applications in a reasonable amount of time demands a high-performance computing environment. Cloud computing is a way of purchasing computing resources on demand through virtualization technologies. It provides the infrastructure to build and run workflow applications, which is called ‘Infrastructure as a Service.' However, it is necessary to schedule workflows on cloud in a way that reduces the cost of leasing resources. Scheduling tasks on resources is a NP hard problem and using meta-heuristic algorithms is an obvious choice for the same. This chapter presents application of nature-inspired algorithms: particle swarm optimization, shuffled frog leaping algorithm and grey wolf optimization algorithm to the workflow scheduling problem on the cloud. Simulation results prove the efficacy of the suggested algorithms.
Cloud computing is a distributed computing paradigm, that provides infrastructure and services to the users using the pay-as-you-use billing model. With the increasing demands and diversity of the scientific workflows, the cloud providers face a fundamental issue of resource provisioning and load balancing. Although, the workflow scheduling in the cloud environment is extensively studied, however, most of the strategies ignore to consider the multiple conflicting objectives of the workflows for scheduling and resource provisioning. To address the above-mentioned issues, in the paper, we introduce a new workflow scheduling strategy using the Firefly algorithm (FA) by considering multiple conflicting objectives including workload of cloud servers, makespan, resource utilization, and reliability. The main purpose of the FA is to find a suitable cloud server for each workflow that can meet its requirements while balancing the loads and resource utilization of the cloud servers. In addition, a rule-based approach is designed to assign the tasks on the suitable VM instances for minimizing the makespan of the workflow while meeting the deadline. The proposed scheduling strategy is evaluated over Google cluster traces using various simulation runs. The control parameters of the FA are also thoroughly investigated for better performance. Through the experimental analysis, we prove that the proposed strategy performs better than the state-of-the-art-algorithms in terms of different Quality-of-Service (QoS) parameters including makespan, reliability, resource utilization and loads of the cloud servers.
Cloud computing is a new paradigm which provides subscription-oriented services. Scheduling the tasks in cloud computing environments is a multi-goal optimization problem, which is NP hard. To exaggerate task scheduling performance and reduce the overall Makespan of the task allocation in clouds, this paper proposes two scheduling algorithms named as TBTS (Threshold based Task scheduling algorithm) and SLA-LB (Service level agreement-based Load Balancing) algorithm. TBTS is two-phase scheduling algorithm which schedules the tasks in a batch. It supports task scheduling in virtual machines with distinct configuration. Furthermore, in TBTS algorithm, threshold data generated based on the ETC (Expected Time to Complete) matrix. Virtual machines which execute tasks with the estimated execution time lesser than threshold value are allocated to the particular task. SLA-LB algorithm is a online model which schedules the task dynamically, based on the requirement of clients, like deadline and budget as the two criteria. Prediction based scheduling is implemented in TBTS to increase the system utilization and to improve the load balancing among the machines by allocation of the minimum configuration machine to the task, based on predicted robust threshold value. SLA-LB uses the level of agreement and finds the required system to reduce the Makespan and increase the cloud-utilization. Simulation of proposed algorithms is performed with benchmark datasets (Braun, 2015) and synthetic datasets are generated with random functions. The proposed TBTS and SLA-LB final values of the proposed algorithms are analyzed with assorted scheduling models, namely SLA-MCT, FCFS, EXIST, LBMM, Lbmaxmin, MINMIN and MAXMIN algorithms. Performance metrics such as Makespan, penalty, gain cost and also the VM utilization factor of proposed algorithm compared with existing algorithms. The comparison analysis among various existing algorithms with TBTS and SLA-LB algorithms show that the proposed methods outperform existing algorithms, even in the scalability situation of the dataset and virtual machines.
Big Data optimization (Big-Opt) refers to optimization problems which require to manage the properties of big data analytics. In the present paper, the Search Manager (SM), a recently proposed framework for hybridizing metaheuristics to improve the performance of optimization algorithms, is extended for multi-objective problems (MOSM), and then five configurations of it by combination of different search strategies are proposed to solve the EEG signal analysis problem which is a member of the big data optimization problems class. Experimental results demonstrate that the proposed configurations of MOSM are efficient in this kind of problems. The configurations are also compared with NSGA-III with uniform crossover and adaptive mutation operators (NSGA-III UCAM), which is a
recently proposed method for Big-Opt problems.
Workflow scheduling is one of the challenging issues in emerging trends of the distributed environment that focuses on satisfying various quality of service (QoS) constraints. The cloud receives the applications as a form of a workflow, consisting of a set of interdependent tasks, to solve the large-scale scientific or enterprise problems. Workflow scheduling in the cloud environment has been studied extensively over the years, and this article provides a comprehensive review of the approaches. This article analyses the characteristics of various workflow scheduling techniques and classifies them based on their objectives and execution model. In addition, the recent technological developments and paradigms such as serverless computing and Fog computing are creating new requirements/opportunities for workflow scheduling in a distributed environment. The serverless infrastructures are mainly designed for processing background tasks such as Internet-of-Things (IoT), web applications, or event-driven applications. To address the ever-increasing demands of resources and to overcome the drawbacks of the cloud-centric IoT, the Fog computing paradigm has been developed. This article also discusses workflow scheduling in the context of these emerging trends of cloud computing.