Dynamic Workflow Research Articles

Multi-Objective Reinforcement Learning Based Algorithm for Dynamic Workflow Scheduling in Cloud Computing

Multi-Objective Reinforcement Learning Based Algorithm for Dynamic Workflow Scheduling in Cloud Computing

Enhancing generalization in genetic programming hyper-heuristics through mini-batch sampling strategies for dynamic workflow scheduling

Genetic Programming Hyper-heuristics (GPHH) have been successfully used to evolve scheduling rules for Dynamic Workflow Scheduling (DWS) as well as other challenging combinatorial optimization problems. The method of sampling training instances has a significant impact on the generalization ability of GPHH, yet they are rarely addressed in existing research. This article aims to fill this gap by proposing a GPHH algorithm with a sampling strategy to thoroughly investigate the impact of six instance sampling strategies on algorithmic generalization, including one rotation strategy, three mini-batch strategies, and two hybrid strategies. Experiments across four scenarios with varying settings reveal that: (1) mini-batch with random sampling can outperform rotation in generalizing to unseen workflow scheduling problems under the same computational cost; (2) employing a hybrid strategy that combines rotation and mini-batch further enhances the generalization ability of GPHH; and (3) mini-batch and hybrid strategies can effectively enable heuristics trained on small-scale training instances generalizing well to large-scale unseen ones. These findings highlight the potential of mini-batch strategies in GPHH, offering improved generalization performance while maintaining diversity and suggesting promising avenues for further exploration in GPHH domains.

Dynamic Workflow Proposal for Continuous Frameless Electromagnetic Neuronavigation in Rigid Neuroendoscopy

Ventriculoscopic neuronavigation has been described in several articles. However, there are different ventriculoscopes and navigation systems. Due to these different combinations, it is difficult to find detailed neuronavigation protocols. We describe, step-by-step, a simple method to navigate both the trajectory until reaching the ventricular system, as well as the intraventricular work. We use a rigid ventriculoscope (LOTTA, KarlStorz) with an electromagnetic stylet (S8-StealthSystem, Medtronic). The protocol is based on a modified or 3-dimensionally printed trocar for navigating the extraventricular step and on a modified pediatric nasogastric tube for the intraventricular navigation. This protocol can be set up in less than 10 minutes. The extraventricular part is navigated by introducing the electromagnetic stylet inside the modified or 3-dimensionally printed trocar. Intraventricular navigation is done by combining a modified pediatric nasogastric tube with the electromagnetic stylet inside the endoscope's working channel. The most critical point is to obtain a blunt-bloodless ventriculostomy while achieving perfect alignment of all targeted structures via pure straight trajectories. This protocol is easy-to-set-up, avoids head rigid-fixation and bulky optical-based attachments to the ventriculoscope, and allows continuous navigation of both parts of the surgery. Since we have implemented this protocol, we have noticed a significant enhancement in both simple and complex ventriculoscopic procedures because the surgery is dramatically simplified.

Approaching Mental Health Through a Preventive Data Analysis Platform

The rising prevalence of chronic diseases and the aging population globally are diminishing the overall quality of life, especially for those with demanding daily routines. As medical advancements extend lifespans, the proportion of individuals over 60 is set to double by 2050, necessitating societal shifts toward health-responsible citizenship. Despite longer lifespans, evidence suggests that older age often accompanies mental health challenges such as anxiety, depression, and substance misuse. Social isolation and loneliness further compound these issues, affecting both physical and mental well-being. Digital wellness empowers individuals to take charge of their health, promoting proactive care and literacy to foster health-conscious citizenship. This paper explores the intersection of mental health, aging populations, preventive wellness initiatives, and health literacy, emphasizing their significance within the Health 5.0 framework, especially for older adults. Traditionally, health regulators offer static workflows for adopting standard procedures in health and well-being, reflecting a reactive approach. However, the evolving landscape of wearable and mobile devices connecting to healthcare IT systems through secure online networks necessitates a shift. Technology now facilitates remote patient monitoring and telemedicine service subscriptions and empowers individuals to manage their health proactively. The BE4YOU project, conducted in Portugal, serves as a case study, facilitating individual maintenance of health and well-being through intelligent and dynamic workflows, which are defined based on analytical models considering each person’s risk profile. By facilitating data sharing, promoting healthy lifestyles, and enabling early detection of mental health issues, the project leverages technology to support personalized monitoring and enhance overall health outcomes. Through technology-driven empowerment, individuals and healthcare professionals are better equipped to assess risks and ensure ongoing wellness monitoring, underscoring the role of technology in fostering personal health and supporting healthcare effectiveness.

A more effective approach for species-level classifications using multi-source remote sensing data: Validation and application to an arid and semi-arid grassland

The distribution of grassland communities significantly influences biodiversity conservation and resource management. Remote sensing datasets, such as Sentinel land cover images with high spatial and temporal resolution, have been commonly employed to simulate the distribution and community structure of plant species in grasslands. Nevertheless, the precision of grassland and community classification requires further enhancement for more effective grassland management. In response, our study presents a dynamic workflow for simulating the distribution of grassland plant species under different hydroclimatic conditions. Leveraging a random forest classifier with multi-source remote sensing features, we mapped the distribution of nine plant communities and six types of land cover in the upstream grasslands of the Xilingol River Basin, Inner Mongolia from 2017 to 2022. Utilizing the Google Earth Engine platform, our findings reveal a remarkable consistency in the overall spatiotemporal accuracy and classification precision of grassland communities across different years, even under varying hydroclimatic conditions. When critical features were carefully selected, the integration of Sentinel-1 (C-band synthetic aperture radar), Sentinel-2 (surface reflectance), SRTM DEM 30 m (Digital Elevation Model data), and MOD16A2.006 (evapotranspiration/latent heat flux product) data yielded a high overall accuracy ranging from 87.0 % to 89.5 % over multiple years. Our study not only determined the spatial patterns of various grassland plant communities but also explored the mechanism of their responses to different hydroclimatic conditions. The proposed approach holds practical implications for fine-scale grassland classification, providing valuable support for landscape classification and monitoring efforts. Furthermore, the insights derived from our findings offer valuable guidance for land managers and policymakers, aiding them in making informed decisions related to grassland conservation and sustainable land use planning.

A Dynamic Workflow Scheduling Method based on MCDM Optimization that Manages Priority Tasks for Fault Tolerance

Because it offers efficient on-demand service delivery over the internet, cloud computing has become more and more popular. An architectural model for cloud computing energy management is provided by the suggested Ant Lion algorithm. Virtual Machines (VMs) in cloud systems are assigned to hosts based not so much on their overall and long-term use, but rather on their immediate resource consumption, including RAM availability. The placement and scheduling processes are frequently computationally demanding and have the potential to affect the performance of deployed virtual machines. In this research work, we offer a strategy that considers the historical resource use of virtual machines (VMs) over time while scheduling them in the cloud. Our goal is to use the Ant lion approach to schedule virtual machines (VMs) in a way that maximizes performance by evaluating the utilization levels of prior VMs. The goal is to reduce the degradation of performance brought about by Cloud administration tasks like as virtual machine deployment, which might impact systems that have already been installed. Furthermore, congested virtual machines (VMs) sometimes take up resources from nearby VMs, increasing the VMs' actual CPU use. Our results show that by learning and adjusting to system behavior over time, our strategy outperforms conventional instant-based physical machine selection. We offer the idea of scheduling virtual machines (VMs) using resource monitoring data from past VM resource use. By using the Ant lion classifier, four fewer physical machines are needed.

PerQueue: managing complex and dynamic workflows.

Workflow managers play a critical role in the efficient planning and execution of complex workloads. A handful of these already exist within the world of computational materials discovery, but their dynamic capabilities are somewhat lacking. The PerQueue workflow manager is the answer to this need. By utilizing modular and dynamic building blocks to define a workflow explicitly before starting, PerQueue can give a better overview of the workflow while allowing full flexibility and high dynamism. To exemplify its usage, we present four use cases at different scales within computational materials discovery. These encapsulate high-throughput screening with Density Functional Theory, using active learning to train a Machine-Learning Interatomic Potential with Molecular Dynamics and reusing this potential for kinetic Monte Carlo simulations of extended systems. Lastly, it is used for an active-learning-accelerated image segmentation procedure with a human-in-the-loop.

Dual-Tree Genetic Programming With Adaptive Mutation for Dynamic Workflow Scheduling in Cloud Computing

Dual-Tree Genetic Programming With Adaptive Mutation for Dynamic Workflow Scheduling in Cloud Computing

Transfer Learning Based Multi-Objective Evolutionary Algorithm for Dynamic Workflow Scheduling in the Cloud

Transfer Learning Based Multi-Objective Evolutionary Algorithm for Dynamic Workflow Scheduling in the Cloud

Resolving the Human-subjects Status of Machine Learning's Crowdworkers

In recent years, machine learning (ML) has relied heavily on crowdworkers both for building datasets and for addressing research questions requiring human interaction or judgment. The diversity of both the tasks performed and the uses of the resulting data render it difficult to determine when crowdworkers are best thought of as workers versus human subjects. These difficulties are compounded by conflicting policies, with some institutions and researchers regarding all ML crowdworkers as human subjects and others holding that they rarely constitute human subjects. Notably few ML papers involving crowdwork mention IRB oversight, raising the prospect of non-compliance with ethical and regulatory requirements. We investigate the appropriate designation of ML crowdsourcing studies, focusing our inquiry on natural language processing to expose unique challenges for research oversight. Crucially, under the U.S. Common Rule, these judgments hinge on determinations of aboutness , concerning both whom (or what) the collected data is about and whom (or what) the analysis is about. We highlight two challenges posed by ML: the same set of workers can serve multiple roles and provide many sorts of information; and ML research tends to embrace a dynamic workflow, where research questions are seldom stated ex ante and data sharing opens the door for future studies to aim questions at different targets. Our analysis exposes a potential loophole in the Common Rule, where researchers can elude research ethics oversight by splitting data collection and analysis into distinct studies. Finally, we offer several policy recommendations to address these concerns.

Assumptions, Perceptions, and Experiences of Behavioral Health Providers Using Telemedicine: Qualitative Study.

The urgent and reactive implementation of telemedicine during the pandemic does not represent a long-term, strategic, and proactive approach to optimizing this technology. The assumptions, perceptions, and experiences of the behavioral health providers using telemedicine can inform system-wide and institutional-level strategies to promote longitudinal maintenance of care delivery, which can reduce the use of high-cost care due to new symptom onset and symptom exacerbation related to service interruptions. We aim to identify the assumptions, perspectives, and experiences of behavioral health clinicians and providers using telemedicine to inform the development of an optimized, sustainable approach to telemedicine implementation. This qualitative study applies the domains of the Consolidated Framework for Implementation Research (CFIR) to structure data collection and analysis from behavioral health providers using telemedicine via an audiovisual connection in the New England region. In total, 12 providers across levels of care were recruited for a 60-minute interview, developed from the CFIR interview guide. Atlas Ti Qualitative Software (version 23; ATLAS.ti Scientific Software Development GmbH) was used to coordinate and facilitate coding among 3 reviewers. Deductive coding was provided from the CFIR interview guide, allowing for data to be categorized by domain and construct. Constructs were analyzed for descriptive themes and tabulated for response frequency. Uncoded data were reviewed and coded in vivo to explore variables contributing to participant perceptions of experience with telemedicine use. Descriptive themes, then analytical themes, were identified. Analytical themes and tabulated frequency of response data were summarized. Finally, a sentiment analysis was completed to derive tone and meaning from the data. Results are reported within the CFIR domains: intervention characteristic, outer setting, inner setting, characteristics of individuals, and process. The findings with ≥90% agreement include "best practice standards were not known"; "telemedicine was believed to be efficient and time-saving for the patient and provider, maximizing productivity and thus increasing access to care"; "telemedicine provided an additional option for patients to access services, promoting sustained continuity and timeliness of care"; "participants did not identify any clear goals related to telemedicine use"; "demonstrated positive affective responses to telemedicine use"; "expressed high efficacy with telemedicine utilization"; and "strong leadership support." These findings support the development of interstate compacts advancing licensure across state lines; payment parity across modalities of care to ensure the financial vitality of behavioral health services; improved dissemination of telehealth training and resources, and telehealth training in academic programs of the health professions; seamless, dynamic workflows to accommodate the changing needs of patient and care continuity; emergency response protocols; and community partnerships to provide private spaces needed for a therapeutic encounter. Future research exploring the patient's experience with telemedicine is needed for all stakeholders to be represented in developing a sustainable, integrated system.

Pandemic Care Through Collaboration: Sharing Lessons from a COVID-19 Police Field Hospital

: In the summer of two years ago in 2021, the prevalence of COVID-19 increased sharply in one of the less privileged provinces. As a result, the number of beds and treatment departments in hospitals, and more importantly, the number of medical specialists in that province, did not meet the existing needs. For this reason, the military healthcare workers decided to come to the aid of that province, considering the organizational mission of the police force. According to the memorandum, the Police Organization, the Governor, and the University of Medical Sciences from that province collaborated to establish a 64-bed field hospital for patients with COVID-19, with two separate wards for male and female patients. The COVID-19 Police Field Hospital provided care for more than a thousand patients over 8 weeks from July 12 to August 31, 2021. In this report, we describe our implementation strategy, including organizational structure, acceptance criteria, and clinical services. Additionally, dynamic workflows require clear lines of relationship, clinical operations expertise, and engaged, interested, and people-friendly healthcare workers.

Dynamic multi-objective workflow scheduling for combined resources in cloud

Cloud resource providers offer idle resources to users as spot instances. The price of the instances changes with market supply and demand, and the dynamic price can have a significant impact on workflow scheduling. In this work, we use a combination of spot and on-demand instances as the foundation cloud resource and characterize the dynamic workflow scheduling problem as a dynamic multi-objective optimization problem (DMOP), where the dynamics originate from the dynamic price of spot instances. The scheduling solution is found by considering three objectives: maximizing the reliability of the instances while minimizing the makespan and cost. In addition, we provide an enhanced MOEAD algorithm called MOEA/D-URDI that combines diversity introduction and uniform random sampling, where the uniform random sampling paradigm is used to generate the initial weight vector. The dynamic multi-objective optimization evolutionary algorithm DMOEA/D-URDI is then created by combining the method with a dynamic optimization framework. Our technique beats existing algorithms, according to experimental data based on dynamic benchmark sets and three well-known scientific procedures in terms of metrics on dynamic benchmark sets and better ensures reliability in scheduling scientific workflows while reducing makespan and cost.

Integrating Cognition Cost With Reliability QoS for Dynamic Workflow Scheduling Using Reinforcement Learning

The rapid rise of microservice architecture poses severe challenges to workflow scheduling, resource allocation, and goal optimization. However, faults and failures usually happen during workflow running. To ensure the workflow's successful execution during scheduling microservices, this paper proposes a dynamic workflow scheduling algorithm by integrating cognition cost and reliability QoS for using reinforcement learning (WS-CCR). Firstly, we explore the ‘restart policy’ of containers in the Kubernetes architecture, which lays the foundation for our work that adopts the redundancy strategy to ensure workflow operation. Then we consider the cognitive cost based on the fact that users have a cognitive process for different microservices in selecting microservices. Additionally, another optimization goal is the reliability of workflows. On this basis, we design a reasonable reward function in reinforcement learning to generate dynamic strategies. Furthermore, following some generated strategies, our engine will schedule candidate microservices for tasks to execute step by step. A series of experiments on Alibaba and business areas workflows have proven the superior performance of our algorithm. Our WS-CCR can generate better Pareto solution sets than other baselines in terms of improving the reliability of running workflows. Finally, we give a case study to prove the practicability of our method.

Genetic Programming for Dynamic Workflow Scheduling in Fog Computing

<bold xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">D</b> ynamic <bold xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">W</b> orkflow <bold xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">S</b> cheduling in <bold xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">F</b> og <bold xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">C</b> omputing (DWSFC) is an important optimisation problem with many real-world applications. The current workflow scheduling problems only consider cloud servers but ignore the roles of mobile devices and edge servers. Some applications need to consider the mobile devices, edge, and cloud servers simultaneously, making them work together to generate an effective schedule. In this article, a new problem model for DWSFC is considered and a new simulator is designed for the new DWSFC problem model. The designed simulator takes the mobile devices, edge, and cloud servers as a whole system, where they all can execute tasks. In the designed simulator, two kinds of decision points are considered, which are the routing decision points and the sequencing decision points. To solve this problem, a new <bold xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">M</b> ulti- <bold xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">T</b> ree <bold xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">G</b> enetic <bold xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">P</b> rogramming (MTGP) method is developed to automatically evolve scheduling heuristics that can make effective real-time decisions on these decision points. The proposed MTGP method with a multi-tree representation can handle the routing decision points and sequencing decision points simultaneously. The experimental results show that the proposed MTGP can achieve significantly better test performance (reduce the makespan by up to 50%) on all the tested scenarios than existing state-of-the-art methods.

An MCDM optimization based dynamic workflow scheduling used to handle priority tasks for fault tolerance in IIOT

Cloud computing's prominence can be mainly attributed to its convenient on-demand service delivery via the internet. The difficulty in managing these requests and allocating them optimally is exacerbated by the growing number of users making them and the variety of criteria used to apply cloud resources. Some of these difficulties include scheduling work while considering the user's Quality of Service (QoS) preferences, ensuring adequate load balancing, and adapting to the ever-changing conditions of the cloud environment. As server capabilities evolve, static load balancers, which distribute requests using fixed ratios of available resources, fail to provide optimum throughput. It has been shown that no other scheduling algorithms offer an adaptive method that considers both load balancing and maximizing the QoS criteria. The authors of this study present a novel adaptive strategy for dealing with this issue by fusing the best-worst multi-criteria decision-making method (BWM) with the compromise ranking method (VIKOR). Task priorities are determined by using the VIKOR method as a decision-maker. Numerical experiments are used to verify the validity of the suggested method, and the methodology is compared to other scheduling methods in terms of a variety of performance criteria. The simulation findings show that the presented method outperforms its rivals regarding throughput, makespan, waiting time, virtual machine (VM) utilization, and VM usage cost across the board of simulated experimental settings.

Technology Focus: Natural Gas Processing and Handling (April 2023)

_ Regardless of which side of the aisle you are on, the critical role that natural gas will play in the energy evolution (or energy transition, as some would say) is undeniable. Natural gas as a fuel is an important bridge toward this evolution and doubles as a reliable partner for renewable energy sources in providing support for wind, solar, and hydroelectric generation by helping to match supply and demand. Renewed interest in, and high-level discussions regarding, the expansion in necessary natural gas infrastructure, the growth in energy consumption, and the expansion of liquefied natural gas business are all strong indicators for technical improvements in the handling and processing of natural gas in this new phase of the energy economy. The first of the case-study-oriented papers in this section highlights a coupled geomechanical and dynamic flow simulation work flow that forms an effective decision-making tool for the process optimization of underground natural gas storage operations in southwest China. Such a multidisciplinary approach allows for the optimization of several key variables to develop strategies for increasing withdrawal rates, meeting the needs of future peak gas demands, and supporting clean-energy initiatives. Indeed, the larger context of underground gas storage plays a key role in addressing supply and demand dynamics as natural gas consumption grows in the future in response to the energy evolution and cleaner-energy initiatives. Another key focus of the energy-evolution discussion is the reduction of carbon emissions through the framework of the circular carbon economy. A case study is presented on the mitigation of routine gas flaring as part of strategic initiatives of the Saudi oil and gas industry. The lessons learned from this case study can be scaled up to the broader industry. In the third highlighted paper, the focus is on the management of methane emissions across the natural gas value chain and a case study on the deployment of methane detection and quantification technologies of the national operator of the Malaysian oil and gas industry. Methane-emissions reduction directly affects advocacy for natural gas as a low-carbon fuel for energy-evolution stakeholders. In the paper, specific methods are developed to quantify methane from fugitive leaks, loss of primary containment, venting and flaring of hydrocarbons, stationary combustion, compressor seals, pneumatic control devices, and pumps. Other than the specific papers highlighted, it is quite encouraging to see cost-effective carbon-emission reduction initiatives at the local operator scale in the sphere of natural gas handling and processing. One such standout that rightfully serves as an example for the entire industry is the 100% elimination of natural-gas-powered pneumatic devices of the largest natural gas producer in the United States, EQT. This meaningful milestone sharply reduced this operator’s annual carbon footprint and, in the wider sense, provided a practical example of aggressively addressing methane emissions through exceptional teamwork and strategy implementation. Recommended additional reading SPE 210890 Strategies for Reducing and Monetizing Fugitive Methane Emissions From Natural Gas Infrastructure by Stefan Diezinger, Siemens Energy SPE 210446 What’s the Best Way To Stabilize Oil in the Permian? An Examination of Different Facilities Layouts by Isabel Chan, TechnipFMC, et al. OTC 31891 A New Approach to Future Floating Liquefied Natural Gas for Offshore Gas Monetization by Xiaogang Liu, Wison Offshore and Marine, et al.

Research on construction data association management of Urban Rail Transit Based on BIM

With the continuous expansion of the scale of urban rail transit construction projects, it is difficult to improve the coordination in construction and the relevance of construction data. In order to solve the problem of data association in construction, an urban rail transit construction data association management system oriented to BIM Technology is proposed. BIM is used to build a dynamic workflow collaboration architecture to manage the data in urban rail transit construction, so as to realize construction collaborative scheduling. In the test and analysis of the system, the comprehensive ability test of the system shows that the minimum response time of the system can be reduced to 1.2 ms. At the same time, the system has ideal visualization effect and obvious practical value in practical application. Therefore, the urban rail transit construction data association management system based on BIM can show ideal results in complex construction environment, which is of great significance to China's urban construction.

A Multi-Objective Based Scheduling Framework for Effective Resource Utilization in Cloud Computing

Cloud computing is a promising platform for running massive workflow applications based on a pay-per-use model. In cloud computing, the reduction of energy consumption and providing security to workflow scheduling are the key research areas. The primary focus of the existing algorithms, viz., particle swarm optimization (PSO), crow Search optimization (CSO) and Other non-metaheuristic algorithms like Round Robbin(RR), SJF, Min-Min, Min-Max etc., is based on the execution time and cost of the workflow applications as a budget constraint. However, these algorithms failed to adequately determine energy consumption, resource utilization, and security in workflow scheduling. To address this issue, a multi-objective scheduling framework is proposed. In this paper, the framework performs dynamic workflow scheduling using universal unique identification- Blake (UUID-Blake), Manhattan Distance-Partition around algorithm (MD-PAM), Linear Scaling-Crow Search Optimization (LS-CSO), Anova-Recurrent Neural Network. The implementation of this framework was achieved in three phases (Phase 1, Phase 2, and Phase 3). Phase 1 is about user registration and authentication using UUID-Blake, which enhances security by allowing legitimate users into the cloud environment. Phase 2 deals with clustering and resource monitoring using MD-PAM and A-RNN, to reduce makespan the similar tasks are clustered using task length and maximize the resource utilization by predicting the resource availability. Phase 3 deals with the scheduling of dynamic workflows using LS-CSO by selecting suitable virtual machines. We have considered the heterogeneous computing scheduling problem (HCSP) and grid workload archive (GWA)-T-12 Bitbrains datasets for comparing our proposed framework with existing works. Based on the result analysis, the proposed LS-SCO outperformed when compared with the algorithms CSO,PSO and RR has achieved better performance.

Multihop Transmission-Oriented Dynamic Workflow Scheduling in Vehicular Cloud

Vehicular cloud as a kind of resource aggregation can provide cloud service flexibly for intelligent transportation systems and individual users. However, the dynamicity of vehicular cloud such as dynamic multihop transmission also brings new challenges to workflow scheduling. In this paper, a heuristic dynamic workflow scheduling (HDWS) strategy is proposed to solve the problem of workflow scheduling with dynamic multihop transmission in vehicular cloud. HDWS establishes a local scheduling based on the current resource status and task sets. The makespan of local scheduling is optimized by reassigning tasks based on task interdependency. At run time of an application, HDWS estimates the makespan in real time based on the current transmission rates and the unready tasks will be rescheduled once it detects the degradation of the makespan. Experimental results show that the proposed HDWS can improve the service success ratio of vehicular cloud and decrease the makespan compared to the existing approaches.