Integrated CloudIoT is an emergingfield of study that integrates the Cloud and the Internet of Things(IoT)to make machines smarter and deal with real-world objects in a distributed manner.It collects data from various ...Integrated CloudIoT is an emergingfield of study that integrates the Cloud and the Internet of Things(IoT)to make machines smarter and deal with real-world objects in a distributed manner.It collects data from various devices and analyses it to increase efficiency and productivity.Because Cloud and IoT are complementary technologies with distinct areas of application,integrating them is difficult.This paper identifies various CloudIoT issues and analyzes them to make a relational model.The Interpretive Structural Modeling(ISM)approach establishes the interrelationship among the problems identified.The issues are categorised based on driving and dependent power,and a hierarchical model is presented.The ISM analysis shows that scheduling is an important aspect and has both(driving and dependence)power to improve the performance of the CloudIoT model.Therefore,existing CloudIoT job scheduling algorithms are ana-lysed,and a cloud-centric scheduling mechanism is proposed to execute IoT jobs on a suitable cloud.The cloud implementation using an open-source framework to simulate Cloud Computing(CloudSim),based on the job’s workload,is pre-sented.Simulation results of the proposed scheduling model indicate better per-formance in terms of Average Waiting Time(AWT)and makespan than existing cloud-based scheduling approaches.展开更多
Modern interactive tools for data analysis and visualisation are designed to expose their functionalities as a service through the Web.We present in this paper a Web API(SWIRRL)that allows Virtual Research Environment...Modern interactive tools for data analysis and visualisation are designed to expose their functionalities as a service through the Web.We present in this paper a Web API(SWIRRL)that allows Virtual Research Environments(VREs)to easily integrate such tools in their websites and re-purpose them to their users.The APl deals,on behalf of the clients,with the underlying complexity of allocating and managing resources within a target cloud platform.By combining storage and containerised services,offering analysis notebooks and other visualisation software,the APl creates dedicated working sessions on-demand,which can be accessed collaboratively.Thanks to the API's support for workflow execution,SWIRRL workspaces can be automatically populated with data of interest collected from external data providers.The system keeps track of updates and changes affecting the data and the tools by adopting versioning and standard provenance technologies.Users are provided with interactive controls enabling traceabilityand recovery actions,including the possibility of creating executable snapshots of their environments.SWIRRL is built in cooperation with two research infrastructures in the field of solid earth science and climate data modeling.We report on the particularadoptions and use cases.展开更多
文摘Integrated CloudIoT is an emergingfield of study that integrates the Cloud and the Internet of Things(IoT)to make machines smarter and deal with real-world objects in a distributed manner.It collects data from various devices and analyses it to increase efficiency and productivity.Because Cloud and IoT are complementary technologies with distinct areas of application,integrating them is difficult.This paper identifies various CloudIoT issues and analyzes them to make a relational model.The Interpretive Structural Modeling(ISM)approach establishes the interrelationship among the problems identified.The issues are categorised based on driving and dependent power,and a hierarchical model is presented.The ISM analysis shows that scheduling is an important aspect and has both(driving and dependence)power to improve the performance of the CloudIoT model.Therefore,existing CloudIoT job scheduling algorithms are ana-lysed,and a cloud-centric scheduling mechanism is proposed to execute IoT jobs on a suitable cloud.The cloud implementation using an open-source framework to simulate Cloud Computing(CloudSim),based on the job’s workload,is pre-sented.Simulation results of the proposed scheduling model indicate better per-formance in terms of Average Waiting Time(AWT)and makespan than existing cloud-based scheduling approaches.
基金supported by the EU H2020 project ENVRIFair(No.824068)ISENES3(No.824084).
文摘Modern interactive tools for data analysis and visualisation are designed to expose their functionalities as a service through the Web.We present in this paper a Web API(SWIRRL)that allows Virtual Research Environments(VREs)to easily integrate such tools in their websites and re-purpose them to their users.The APl deals,on behalf of the clients,with the underlying complexity of allocating and managing resources within a target cloud platform.By combining storage and containerised services,offering analysis notebooks and other visualisation software,the APl creates dedicated working sessions on-demand,which can be accessed collaboratively.Thanks to the API's support for workflow execution,SWIRRL workspaces can be automatically populated with data of interest collected from external data providers.The system keeps track of updates and changes affecting the data and the tools by adopting versioning and standard provenance technologies.Users are provided with interactive controls enabling traceabilityand recovery actions,including the possibility of creating executable snapshots of their environments.SWIRRL is built in cooperation with two research infrastructures in the field of solid earth science and climate data modeling.We report on the particularadoptions and use cases.
