Article
Computer Science, Information Systems
Meng Xu, Yi Mei, Shiqiang Zhu, Beibei Zhang, Tian Xiang, Fangfang Zhang, Mengjie Zhang
Summary: Dynamic Workflow Scheduling in Fog Computing is a significant optimization problem that involves the coordination of cloud servers, mobile devices, and edge servers. This article proposes a new problem model and simulator, as well as a Multi-Tree Genetic Programming method to address the problem. Experimental results demonstrate that the proposed method achieves significantly better performance across all tested scenarios.
IEEE TRANSACTIONS ON SERVICES COMPUTING
(2023)
Article
Automation & Control Systems
Lingjuan Ye, Yuanqing Xia, Liwen Yang, Yufeng Zhan
Summary: This paper investigates the problem of stochastic multi-workflows scheduling in clouds and proposes an efficient algorithm called SMWDSA to optimize the scheduling cost. The algorithm includes three stages and achieves the goal of meeting workflow deadline constraints and minimizing workflow scheduling cost. The experimental results demonstrate the superiority of SMWDSA compared to other algorithms.
IEEE TRANSACTIONS ON AUTOMATION SCIENCE AND ENGINEERING
(2022)
Article
Computer Science, Information Systems
Matias Hirsch, Cristian Mateos, Alejandro Zunino, Tim A. Majchrzak, Tor-Morten Gronli, Hermann Kaindl
Summary: Although the computing resources of today's smartphones are often underutilized, leveraging them could have significant benefits in edge and fog computing environments. Challenges arise, such as job scheduling, due to differences in computational capabilities and energy usage among individual devices. Research explores task execution schemes relying on clusters of mobile devices, evaluating practical heuristics for job scheduling with simulated scenarios and real-time stream processing using Tensorflow models on mid-range and low-end smartphones in small clusters. Ultimately, the goal is to improve task scheduling for dew computing contexts and enhance citizen participation by harnessing dew computing resources in urban environments for smarter city development.
Article
Computer Science, Hardware & Architecture
Dongsheng Li, Zhiyao Hu, Zhiquan Lai, Yiming Zhang, Kai Lu
Summary: This research introduces a new scheduling unit abstraction called coBranch, which considers the dependencies between computation stages and coflows to jointly schedule coflows and jobs, aiming to reduce the average job completion time while increasing inter-job parallelism. By employing an urgency-based mechanism, the proposed method achieved significant reductions in average JCT and outperformed existing scheduling methods in prototype-based experiments and large-scale simulations.
IEEE TRANSACTIONS ON COMPUTERS
(2021)
Article
Computer Science, Information Systems
Yuanjun Laili, Fuqiang Guo, Lei Ren, Xiang Li, Yulin Li, Lin Zhang
Summary: Industrial Internet of Things (IIoT) is becoming intelligent with large-scale collaborative cloud and edge resources, enabling online supervision, fast analysis, and precise control for manufacturing job shops. However, processing large-scale industrial computation online leads to significant communication overhead and energy consumption among cloud, edge, and end devices. To enhance cloud-edge collaboration, this article proposes a practical task scheduling model considering two types of collaborative modes. A parallel group-merge evolutionary algorithm is introduced to assign thousands of tasks within seconds, by dividing them into weakly correlated groups and applying modified evolutionary operators for finding a subsolution for each group. Experimental results demonstrate that this method can achieve swift allocation of tasks to cloud and edge servers, reducing overall task computing time by 36.97% and saving up to 23.71% of energy consumption.
IEEE INTERNET OF THINGS JOURNAL
(2023)
Article
Automation & Control Systems
Su Nguyen, Mengjie Zhang, Damminda Alahakoon, Kay Chen Tan
Summary: A novel people-centric evolutionary system for dynamic production scheduling has been developed with new techniques and models to enhance the efficiency of genetic programming, outperforming existing algorithms in dynamic flexible job shop scheduling experiments.
IEEE TRANSACTIONS ON CYBERNETICS
(2021)
Article
Computer Science, Interdisciplinary Applications
Shashank Srivastava, Sandeep Saxena, Rajkumar Buyya, Manoj Kumar, Achyut Shankar, Bharat Bhushan
Summary: The paper focuses on the advancement of a cloud administration's provisioning structure by establishing a dynamic load-balancer for the cloud. It proposes a framework and protocol using Distributed Hash Table (DHT) and gossip protocol for resource management and load balancing. The protocol is adaptable, reliable, and scalable, supporting green computing.
SIMULATION MODELLING PRACTICE AND THEORY
(2021)
Article
Computer Science, Hardware & Architecture
Rongli Chen, Xiaozhong Chen, Cairu Yang
Summary: The rapid development of internet technology has led to the growth of data centers, which consume massive amounts of power leading to significant carbon emissions. The issue of energy consumption has become a crucial topic in cloud computing research, with the proposed energy-saving job-scheduling method showing promising results in reducing energy consumption and job discard rate. The method considers task dependency in a cloud computing environment and models energy consumption based on the frequency and kernel number of the virtual machine CPU.
JOURNAL OF SUPERCOMPUTING
(2022)
Article
Computer Science, Artificial Intelligence
Youyou Kang, Li Pan, Shijun Liu
Summary: This paper proposes an online scheduling algorithm to optimize job scheduling decisions for big data analysis, improving job execution efficiency and economic benefits; theoretical analysis and extensive simulation experiments demonstrate that the algorithm achieves more stable performance compared to the traditional two-phase scheduling algorithm.
KNOWLEDGE-BASED SYSTEMS
(2022)
Article
Computer Science, Information Systems
Yejian Zhao, Yanhong Wang, Yuanyuan Tan, Jun Zhang, Hongxia Yu
Summary: Jobshop scheduling is a classic instance in production scheduling, and the proposed scheduling algorithm based on DQN is more effective in improving the market competitiveness of manufacturing enterprises.
Article
Computer Science, Information Systems
Xiaoyong Tang, Yi Liu, Zeng Zeng, Bharadwaj Veeravalli
Summary: Nowadays, increasing number of services are provided to individuals and organizations through cloud computing systems in a pay-as-you-use model. This paper proposes a cloud computing systems resources management architecture and a CUDA-enabled parallel two-dimensional long short-term memory neural network to predict software faults in cloud VMs. It also introduces an effective primary/backup cloud service cost calculation approach and a reliability and cost aware job scheduling algorithm. Experimental results show that the proposed algorithm outperforms other algorithms in terms of average service cost and rejection rate and is suitable for cloud services with high reliability and low-cost requirements.
IEEE TRANSACTIONS ON CLOUD COMPUTING
(2023)
Review
Computer Science, Theory & Methods
Youyou Kang, Li Pan, Shijun Liu
Summary: This article summarizes the research work on big data analytics job scheduling in cloud environments, categorizes scheduling algorithms based on different research focuses, compares the advantages and disadvantages of existing research, and points out the directions and challenges for future research.
FUTURE GENERATION COMPUTER SYSTEMS-THE INTERNATIONAL JOURNAL OF ESCIENCE
(2022)
Article
Computer Science, Information Systems
Tiago A. O. Alves, Leandro A. J. Marzulo, Sandip Kundu, Felipe M. G. Franca
Summary: This paper presents techniques for performing concurrency analysis on generic dynamic dataflow graphs, even in the presence of cycles, allowing concurrency between different iterations. The novelty of the approach lies in allowing concurrency between nodes from different loops that can be proven independent. The paper provides theoretical tools for obtaining bounds and demonstrates the implementation of parallel dataflow runtime on representative graphs to compare performance against derived bounds.
IEEE TRANSACTIONS ON EMERGING TOPICS IN COMPUTING
(2021)
Article
Automation & Control Systems
Shunmei Meng, Weijia Huang, Xiaochun Yin, Mohammad R. Khosravi, Qianmu Li, Shaohua Wan, Lianyong Qi
Summary: This article proposes a security-aware dynamic scheduling method for real-time resource allocation in ICS. A three-level security model is designed for tasks and cloud resources in ICS, and a two-tier heterogeneous cloud architecture is introduced. A security-aware scheduling method based on distributed particle swarm optimization is presented for resource allocation, with a dynamic scheduling mechanism based on dynamic workflow model to address the dynamics of edge resources and mobile industrial applications.
IEEE TRANSACTIONS ON INDUSTRIAL INFORMATICS
(2021)
Article
Computer Science, Information Systems
Xiao Ma, Ao Zhou, Shan Zhang, Qing Li, Alex X. Liu, Shangguang Wang
Summary: The cloud-assisted mobile edge computing system is an important architecture for processing computation-intensive and delay-sensitive mobile applications efficiently. The paper proposes a Water-filling Based Dynamic Task Scheduling algorithm to solve the dynamic task scheduling problem, aiming at minimizing average task response time within the resource budget limit.
IEEE TRANSACTIONS ON MOBILE COMPUTING
(2023)