Mobile edge computing(MEC) is a cloud server running at the edge of a mobile network, which can effectively reduce network communication delay. However, due to the numerous edge servers and devices in the MEC, there m...Mobile edge computing(MEC) is a cloud server running at the edge of a mobile network, which can effectively reduce network communication delay. However, due to the numerous edge servers and devices in the MEC, there may be multiple servers and devices that can provide services to the same user simultaneously. This paper proposes a userside adaptive user service deployment algorithm ASD(Adaptive Service Deployment) based on reinforcement learning algorithms. Without relying on complex system information, it can master only a few tasks and users. In the case of attributes, perform effective service deployment decisions, analyze and redefine the key parameters of existing algorithms, and dynamically adjust strategies according to task types and available node types to optimize user experience delay. Experiments show that the ASD algorithm can implement user-side decision-making for service deployment. While effectively improving parameter settings in the traditional Multi-Armed Bandit algorithm,it can reduce user-perceived delay and enhance service quality compared with other strategies.展开更多
Geo-analysis models can be shared and reused via model-services to support more effective responses to risks and help to build a sustainable world.The deployment of model-services typically requires significant effort...Geo-analysis models can be shared and reused via model-services to support more effective responses to risks and help to build a sustainable world.The deployment of model-services typically requires significant effort,primarily because of the complexity and disciplinary specifics of geo-analysis models.Various modelling participants engage in the collaborative modelling process:geo-analysis model resources are provided by model providers,computational resources are provided by computational resource providers,and the published model-services are accessed by model users.This paper primarily focuses on model-service deployment,with the basic goal of providing a collaboration-oriented method for modelling participants to conveniently work together and make full use of modelling and computational resources across an open web environment.For model resource providers,a model-deployment description method is studied to help build model-deployment packages;for computational resource providers,a computational resource description method is studied to help build model-service containers and connectors.An experimental system for sharing and reusing geo-analysis models is built to verify the capability and feasibility of the proposed methods.Through this strategy,modellers from dispersed regions can work together more easily,thus providing dynamic and reliable geospatial information for Future Earth studies.展开更多
The integration of cloud and IoT edge devices is of significance in reducing the latency of IoT stream data processing by moving services closer to the edge-end.In this connection,a key issue is to determine when and ...The integration of cloud and IoT edge devices is of significance in reducing the latency of IoT stream data processing by moving services closer to the edge-end.In this connection,a key issue is to determine when and where services should be deployed.Common service deployment strategies used to be static based on the rules defined at the design time.However,dynamically changing IoT environments bring about unexpected situations such as out-of-range stream fluctuation,where the static service deployment solutions are not efficient.In this paper,we propose a dynamic service deployment mechanism based on the prediction of upcoming stream data.To effectively predict upcoming workloads,we combine the online machine learning methods with an online optimization algorithm for service deployment.A simulation-based evaluation demonstrates that,compared with those state-of-the art approaches,the approach proposed in this paper has a lower latency of stream processing.展开更多
Cloud computing makes computing power universally available and provides flexibility in resource acquisition. It allows for scalable provision of services and more reasonable use of resources. This article considers c...Cloud computing makes computing power universally available and provides flexibility in resource acquisition. It allows for scalable provision of services and more reasonable use of resources. This article considers cloud service deployment and virtualization from the perspective of mobile operators. A solution is proposed that allows mobile operators to maximize profits with minimal investment,展开更多
基金supported in part by the Industrial Internet Innovation and Development Project "Industrial robot external safety enhancement device"(TC200H030)the Cooperation project between Chongqing Municipal undergraduate universities and institutes affiliated to CAS (HZ2021015)
文摘Mobile edge computing(MEC) is a cloud server running at the edge of a mobile network, which can effectively reduce network communication delay. However, due to the numerous edge servers and devices in the MEC, there may be multiple servers and devices that can provide services to the same user simultaneously. This paper proposes a userside adaptive user service deployment algorithm ASD(Adaptive Service Deployment) based on reinforcement learning algorithms. Without relying on complex system information, it can master only a few tasks and users. In the case of attributes, perform effective service deployment decisions, analyze and redefine the key parameters of existing algorithms, and dynamically adjust strategies according to task types and available node types to optimize user experience delay. Experiments show that the ASD algorithm can implement user-side decision-making for service deployment. While effectively improving parameter settings in the traditional Multi-Armed Bandit algorithm,it can reduce user-perceived delay and enhance service quality compared with other strategies.
基金supported by the National Basic Research Program of China(973 Program)under Grant number 2015CB954102the National Natural Science Foundation of China under Grant number 41471317,Grant number 41301414 and Grant number 41371424the Priority Academic Program Development of the Jiangsu Higher Education Institutions under Grant number 164320H116.
文摘Geo-analysis models can be shared and reused via model-services to support more effective responses to risks and help to build a sustainable world.The deployment of model-services typically requires significant effort,primarily because of the complexity and disciplinary specifics of geo-analysis models.Various modelling participants engage in the collaborative modelling process:geo-analysis model resources are provided by model providers,computational resources are provided by computational resource providers,and the published model-services are accessed by model users.This paper primarily focuses on model-service deployment,with the basic goal of providing a collaboration-oriented method for modelling participants to conveniently work together and make full use of modelling and computational resources across an open web environment.For model resource providers,a model-deployment description method is studied to help build model-deployment packages;for computational resource providers,a computational resource description method is studied to help build model-service containers and connectors.An experimental system for sharing and reusing geo-analysis models is built to verify the capability and feasibility of the proposed methods.Through this strategy,modellers from dispersed regions can work together more easily,thus providing dynamic and reliable geospatial information for Future Earth studies.
基金supported by the General Program of National Natural Science Fouddation of China:Analytical Method Reserach of Loop and Recursion(No.61872262/F020106)the Key Project of the National Natural Science Foundation of China:Research on Big Service Theory and Methods in Big Data Environment(No.61832004).
文摘The integration of cloud and IoT edge devices is of significance in reducing the latency of IoT stream data processing by moving services closer to the edge-end.In this connection,a key issue is to determine when and where services should be deployed.Common service deployment strategies used to be static based on the rules defined at the design time.However,dynamically changing IoT environments bring about unexpected situations such as out-of-range stream fluctuation,where the static service deployment solutions are not efficient.In this paper,we propose a dynamic service deployment mechanism based on the prediction of upcoming stream data.To effectively predict upcoming workloads,we combine the online machine learning methods with an online optimization algorithm for service deployment.A simulation-based evaluation demonstrates that,compared with those state-of-the art approaches,the approach proposed in this paper has a lower latency of stream processing.
文摘Cloud computing makes computing power universally available and provides flexibility in resource acquisition. It allows for scalable provision of services and more reasonable use of resources. This article considers cloud service deployment and virtualization from the perspective of mobile operators. A solution is proposed that allows mobile operators to maximize profits with minimal investment,
