Search

Heterogeneous cloud datacenters are well-suited and cost-efficient platforms for execution of scientific workflows requested from academics. Workflow scheduling algorithms have drastic impacts on the objectives that stakeholders in the system expect. This paper models the scientific workflow scheduling issue to a bi-objective optimization problem with makespan and reliability optimization approach because the users not only expect to have quick response, but also they need reliable executions. To address the issue, a new system framework and different concepts are introduced. A centralized log as a repository module is embedded in the system framework to register all kinds of system failures. In addition to, the new scheduling failure factor (SFF), which has reciprocal relation with system reliability, is defined. Therefore, the broker module quantifies the failure proneness of all resources and the most reliable ones are incorporated in scheduling model. The aforementioned scheduling model is then formulated to a bi-objective optimization problem with makespan and SFF minimization viewpoint which is an NP-Hard problem. To solve this combinatorial problem, a hybrid bi-objective discrete cuckoo search algorithm (HDCSA) is proposed. The proposed hybrid algorithm utilizes different novel Levy flight operators commensurate with discrete search space that makes good balance between exploration and exploitation in optimization process. The proposed HDCSA was validated in 12 extensive scenarios that were conducted on both symmetric and asymmetric scientific workflows in different conditions. The final results prove that the proposed bi-objective HDCSA scheduler has the amount of 22.11%, 12.97%, 11.81%, 12.18%, and 12.42% on average improvement against other state-of-the-arts in terms of makespan, SFF, speedup, efficiency, and SLR, respectively, which are prominent performance evaluation metrics is this scheduling domain. [ABSTRACT FROM AUTHOR]

The Internet of Things (IoT) is a telecommunication network in the next generation of applications with the rapid progress of wireless sensor network techniques that have touched many spheres of life today. Hardware, telephony, communications, storage, secure platforms, software and services, and data processing platforms are all part of the IoT environment. IoT sensors collect data from their environment and share it by connecting to the Internet gateway. These sensors often perform tasks without human intervention. This article aims to review real-time scheduling in the IoT to fully understand the issues raised in this area published from 2018 to 2022. A classification for IoT applications based on practical application is provided for selected studies. Selected studies include healthcare, infrastructure, industrial applications, smart city, commercial applications, environmental protection, and general IoT applications. Studies are sorted into groups based on related applications and compared based on indicators such as performance time, energy consumption, makespan, and assessment environments depending on the provided classification. Finally, this paper discusses all reviewed studies' main concepts, disadvantages, advantages, and future work. [ABSTRACT FROM AUTHOR]

Today, the Internet of Things (IoT) technology is increasingly applied to various applications such as electronic health, smart cities, car networks, positioning systems, and virtual reality. Due to this growing use of IoT, huge amounts of data are being generated unceasingly, which require extensive and continuous processing. As IoT sensors generally have resource constraints, and also the distance between these sensors and cloud data centers is so large, using the cloud is not a good solution to overcome this constraint, particularly in performing tasks that are sensitive to latency. To overcome these problems, fog computing is used as an intermediate layer between IoT sensors and cloud data centers, which provides resources at the edge of the network for IoT applications. One of the main issues in the fog–cloud environment is choosing the right place to perform tasks based on available resources, which requires a proper scheduling. To address this challenge, the present paper proposes a task scheduling model based on the Task Possible Execution Level (TPEL) in the fog–cloud environment to minimize delay and energy consumption. In TPEL, first, tasks are prioritized depending on their deadlines and resources; then, the possible execution levels of tasks are determined. Finally, the most appropriate place to perform the task is selected from the possible execution levels based on the integer linear programming model. The simulation results showed that the proposed TPEL is more efficient by 28.6% in terms of delay, 49.3% in terms of energy consumption, 25.5% in terms of scheduling time, and 51.8% in terms of response time compared to First-Come, First-Served, Round Robin, Algorithm Non-dominated Sorting Genetic Algorithm II, and Delay Aware Scheduling algorithm. [ABSTRACT FROM AUTHOR]

In recent years, cloud computing can be considered an emerging technology that can share resources with users. Because cloud computing is on-demand, efficient use of resources such as memory, processors, bandwidth, etc., is a big challenge. Despite the advantages of cloud computing, sometimes it is not a proper choice due to its delay in responding appropriately to existing requests, which led to the need for another technology called fog computing. Fog computing reduces traffic and time lags by expanding cloud services to the network and closer to users. It can schedule resources with higher efficiency and utilize them to impact the user's experience dramatically. This paper aims to survey some studies that have been done in the field of scheduling in fog/cloud computing environments. The focus of this survey is on published studies between 2015 and 2021 in journals or conferences. We selected 71 studies in a systematic literature review (SLR) from four major scientific databases based on their relation to our paper. We classified these studies into five categories based on their traced parameters and their focus area. This classification comprises 1—performance 2—energy efficiency, 3—resource utilization, 4—performance and energy efficiency, and 5—performance and resource utilization simultaneously. 42.3% of the studies focused on performance, 9.9% on energy efficiency, 7.0% on resource utilization, 21.1% on both performance and energy efficiency, and 19.7% on both performance and resource utilization. Finally, we present challenges and open issues in the resource scheduling methods in fog/cloud computing environments. [ABSTRACT FROM AUTHOR]

Given the significant development of the Internet of Things (IoT) in recent years as well as the growing need for data around the world, cloud computing alone is not able to manage this volume of data. Accordingly, fog computing has just now become a popular paradigm for further data analysis in close proximity to devices generating and processing data instantly, in order to solve various problems of existing cloud-only based systems. With regard to the complexity and the wide variety of types of computational resources such as cloud servers and fog nodes, workflow scheduling is thus one of the most important challenges in fog computing environments. To address such a problem, this paper presents software architecture for scheduling multiple workflows in cloud-fog environments simultaneously. Within this scheduling, workflow clustering and priority of workflows are also taken into account. As well, architecture layers, components, as well as their major interactions are represented using 4 + 1 architectural view models. The architecture components are ultimately proposed to meet quality attributes such as availability, reliability, recoverability, interoperability, and performance. The proposed architecture evaluation is based on the Architecture Tradeoff Analysis Method (ATAM) is a scenario-based technique. Compared with previous works, various scenarios and more quality attributes are discussed within this evaluation in addition to clustering and prioritizing workflows. [ABSTRACT FROM AUTHOR]

Cloud providers deliver heterogeneous virtual machines to run complicated jobs submitted by users. The task scheduling issue is formulated to a discrete optimization problem which is well-known NP-Hard. This paper presents a hybrid meta-heuristic algorithm based on genetic and thermodynamic simulated annealing algorithms to solve this problem. In the proposed algorithm, the genetic and simulated annealing algorithms have respective global and local search inclinations covering each other's shortcomings. A novel theorem is presented and applied to produce a semi-conducted initial population. In a used genetic algorithm with a global trend, the crossover operator is performed to explore search space. The thermodynamic simulated annealing algorithm is utilized to improve the efficiency, which considers entropy and energy difference concepts in the cooling schedule process. After obtaining a suitable solution, one of the three novel neighbor operators is randomly called to enhance the given solution potentially. In this way, the efficient balance between exploration and exploitation in the search space is achieved. Simulation results prove that the proposed hybrid algorithm has 10.17%, 9.31%, 7.76%, and 8.21% dominance in terms of makespan, schedule length ratio, speedup, and efficiency against other comparative algorithms. [ABSTRACT FROM AUTHOR]

For the Internet of Things (IoT) that have low processing power, fog computing can play an essential role as a processing resource to execute their tasks. Timely execution of the service and the optimal service cost are two major challenges of task scheduling in fog computing. In this article, a privacy architecture is proposed for task scheduling in IoT, and based on this architecture, a multi-objective algorithm is presented to minimize the service time and service cost. We simulate the problem with four scenarios: easy, medium, semi-hard, and hard, and then compare it with other multi-objective algorithms. Since the proposed algorithm is multi-objective, the goal programming approach (GPA) is used to choose one possible solution. The simulation results show that our proposed algorithm has better performance and higher convergence speed to the optimal solution than other algorithms while considering the privacy requirements of IoT devices. [ABSTRACT FROM AUTHOR]

Fog computing is an interesting technology aimed at providing various processing and storage resources at the IoT networks' edge. Energy consumption is one of the essential factors that can directly impact the maintenance cost and CO 2 emissions of fog environments. Energy consumption can be mitigated by effective scheduling approaches, in which tasks are going to be mapped on the best possible resources regarding some conflicting objectives. To deal with these issues, we introduce an opposition-based hybrid discrete optimization algorithm, called DMFO-DE. For this purpose, first, a discrete and Opposition-Based Learning (OBL) version of the Moth–Flame Optimization (MFO) algorithm is provided, and it then is combined with the Differential Evolution (DE) algorithm to improve the convergence speed and prevent local optima problem. The DMFO-DE is then employed for scientific workflow scheduling in fog computing environments using the Dynamic Voltage and Frequency Scaling (DVFS) method. The Heterogeneous Earliest Finish Time (HEFT) algorithm is used to find the tasks execution order in the scientific workflows. Our workflow scheduling approach mainly tries to decrease the scheduling process's energy consumption by minimizing the applied Virtual Machines (VMs), makespan, and communication between dependent tasks. For evaluating the performance of the proposed scheduling scheme, extensive simulations are conducted on the scientific workflows with four different sizes. The experimental results indicate that scheduling using the DMFO-DE algorithm can outperform other metrics such as the number of applied VMs, and energy consumption. • Presenting DMFO a discrete version of the MFO algorithm. • Presenting DMFO-DE, a hybrid and OBL-based optimization algorithm. • Introducing a scientific workflow scheduling scheme using the DMFO-DE. • Effective energy management using DVFS. • Extensive simulations using the iFogSim simulator tool. [ABSTRACT FROM AUTHOR]