Multi-cloud computing promises to deliver certain benefits, like performance optimisation and cost reduction. However, most cloud application modelling languages are tight to one cloud platform and do not have the right expressiveness to cover all application lifecycle phases. This survey attempts to review the most important from these languages, which facilitate application provisioning in commercial cloud platforms. The main review goals are to: (a) highlight those languages already or nearly multi-cloud enabled; (b) determine those parts of the remaining languages that must be extended to support multi-cloud application modelling. This review results also lead to drawing some future work directions towards producing an ideal multi-cloud application specification language.
{"title":"Are Cloud Modelling Languages Ready for Multi-Cloud?","authors":"K. Kritikos, Pawel Skrzypek","doi":"10.1145/3368235.3368840","DOIUrl":"https://doi.org/10.1145/3368235.3368840","url":null,"abstract":"Multi-cloud computing promises to deliver certain benefits, like performance optimisation and cost reduction. However, most cloud application modelling languages are tight to one cloud platform and do not have the right expressiveness to cover all application lifecycle phases. This survey attempts to review the most important from these languages, which facilitate application provisioning in commercial cloud platforms. The main review goals are to: (a) highlight those languages already or nearly multi-cloud enabled; (b) determine those parts of the remaining languages that must be extended to support multi-cloud application modelling. This review results also lead to drawing some future work directions towards producing an ideal multi-cloud application specification language.","PeriodicalId":166357,"journal":{"name":"Proceedings of the 12th IEEE/ACM International Conference on Utility and Cloud Computing Companion","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-12-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130382727","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Clouds offer significant advantages over traditional cluster computing architectures including flexibility, high-availability, ease of deployment, and on-demand resource allocation -- all packed up in an attractive pay-as-you-go economic model for the users. However, cloud users are often forced into vendor lock-in due to the use of incompatible APIs, cloud-specific services, and complex pricing models of the cloud service providers (CSPs). Cloud management platforms (CMPs), supporting hybrid and multi-cloud deployment, offer an answer by providing a unified abstract interface to multiple cloud platforms. Nonetheless, modelling applications to use multi-clouds, automated resource selection based on user requirements from various available CSPs, cost optimization, security, and runtime adaptation of deployed applications still remain a challenge. In this tutorial, we provide a practical introduction to multi-cloud application modelling, configuration, deployment, and adaptation. We survey existing CMPs, compare their features and modelling methods. Finally, we provide a practical hands-on training for getting your applications ready for multi-cloud using selected tools. By the end of this tutorial, attendees should be able to understand the benefits of the multi-cloud approach, and prepared to deploy their first managed multi-cloud application.
{"title":"Good Bye Vendor Lock-in: Getting your Cloud Applications Multi-Cloud Ready!","authors":"M. Rózanska, K. Kritikos","doi":"10.1145/3368235.3370267","DOIUrl":"https://doi.org/10.1145/3368235.3370267","url":null,"abstract":"Clouds offer significant advantages over traditional cluster computing architectures including flexibility, high-availability, ease of deployment, and on-demand resource allocation -- all packed up in an attractive pay-as-you-go economic model for the users. However, cloud users are often forced into vendor lock-in due to the use of incompatible APIs, cloud-specific services, and complex pricing models of the cloud service providers (CSPs). Cloud management platforms (CMPs), supporting hybrid and multi-cloud deployment, offer an answer by providing a unified abstract interface to multiple cloud platforms. Nonetheless, modelling applications to use multi-clouds, automated resource selection based on user requirements from various available CSPs, cost optimization, security, and runtime adaptation of deployed applications still remain a challenge. In this tutorial, we provide a practical introduction to multi-cloud application modelling, configuration, deployment, and adaptation. We survey existing CMPs, compare their features and modelling methods. Finally, we provide a practical hands-on training for getting your applications ready for multi-cloud using selected tools. By the end of this tutorial, attendees should be able to understand the benefits of the multi-cloud approach, and prepared to deploy their first managed multi-cloud application.","PeriodicalId":166357,"journal":{"name":"Proceedings of the 12th IEEE/ACM International Conference on Utility and Cloud Computing Companion","volume":"38 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-12-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125108168","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
For production line level analysis such as bottleneck analysis, distributed manufacturing data must be integrated and cleansed according to domain knowledge. As a result, these data are normally analysed by experts, and could be time-consuming. Although many methods have been proposed to support such progress, there is still no system designed for users without detailed domain knowledge. Hence, we develop a comprehensive data processing platform automatically integrating and cleansing raw manufacturing data according to analysis context specified by user. By applying our proposal to different scenarios and products in a real-life automotive parts factory, we illustrate that the platform can be used by anyone with basic knowledge about target product. As a result, various forms of value can be extracted by even ordinary workers.
{"title":"Towards a Comprehensive Data Processing Platform","authors":"Q. Xiu, Keiro Muro","doi":"10.1145/3368235.3369373","DOIUrl":"https://doi.org/10.1145/3368235.3369373","url":null,"abstract":"For production line level analysis such as bottleneck analysis, distributed manufacturing data must be integrated and cleansed according to domain knowledge. As a result, these data are normally analysed by experts, and could be time-consuming. Although many methods have been proposed to support such progress, there is still no system designed for users without detailed domain knowledge. Hence, we develop a comprehensive data processing platform automatically integrating and cleansing raw manufacturing data according to analysis context specified by user. By applying our proposal to different scenarios and products in a real-life automotive parts factory, we illustrate that the platform can be used by anyone with basic knowledge about target product. As a result, various forms of value can be extracted by even ordinary workers.","PeriodicalId":166357,"journal":{"name":"Proceedings of the 12th IEEE/ACM International Conference on Utility and Cloud Computing Companion","volume":"30 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-12-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134222252","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
As a result of the many technical advances in microcomputers and mobile connectivity, the Internet of Things (IoT) has been on the rise in the recent decade. Due to the broad spectrum of applications, networks facilitating IoT scenarios can be of very different scale and complexity. Additionally, connected devices are uncommonly heterogeneous, including micro controllers, smartphones, fog nodes and server infrastructures. Therefore, testing IoT applications is difficult, motivating adequate tool support. In this paper, we present Héctor, a framework for the automatic testing of IoT applications. Héctor allows the automated execution of user-defined experiments on agnostic IoT testbeds. To test applications independently of the availability of required devices, the framework is able to generate virtual testbeds with adjustable network properties. Our evaluations show that simple experiments can be easily automated across a broad spectrum of testbeds. However, the results also indicate that there is considerable interference in experiments, in which many devices are emulated, due to the high resource demand of system emulation.
{"title":"Héctor: A Framework for Testing IoT Applications Across Heterogeneous Edge and Cloud Testbeds","authors":"Ilja Behnke, L. Thamsen, O. Kao","doi":"10.1145/3368235.3368832","DOIUrl":"https://doi.org/10.1145/3368235.3368832","url":null,"abstract":"As a result of the many technical advances in microcomputers and mobile connectivity, the Internet of Things (IoT) has been on the rise in the recent decade. Due to the broad spectrum of applications, networks facilitating IoT scenarios can be of very different scale and complexity. Additionally, connected devices are uncommonly heterogeneous, including micro controllers, smartphones, fog nodes and server infrastructures. Therefore, testing IoT applications is difficult, motivating adequate tool support. In this paper, we present Héctor, a framework for the automatic testing of IoT applications. Héctor allows the automated execution of user-defined experiments on agnostic IoT testbeds. To test applications independently of the availability of required devices, the framework is able to generate virtual testbeds with adjustable network properties. Our evaluations show that simple experiments can be easily automated across a broad spectrum of testbeds. However, the results also indicate that there is considerable interference in experiments, in which many devices are emulated, due to the high resource demand of system emulation.","PeriodicalId":166357,"journal":{"name":"Proceedings of the 12th IEEE/ACM International Conference on Utility and Cloud Computing Companion","volume":"49 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-12-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115901387","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Hiroki Kawahara, R. Yamamoto, S. Ohzahata, Toshihiko Kato
Overlay network virtualization provides virtual networks over IP networks. Building virtual networks over IP networks achieves high mobility to realize a multi-tenant network. Although the encapsulation protocol, such as VXLAN, is widely used to implement overlay network virtualization, the encapsulation process degrades throughput. Therefore, we propose to combine NoEncap and Hoverboard model to improve throughput in commodity devices. NoEncap achieves performance enhancement of overlay network virtualization for commodity devices using non-encapsulation but increases the control plane overhead. The Hoverboard model reduces the control plane overhead. This paper presents the evaluation of the throughput enhancement by the proposed method.
{"title":"Throughput Enhancement with Overlay Network Virtualization Using Commodity Devices","authors":"Hiroki Kawahara, R. Yamamoto, S. Ohzahata, Toshihiko Kato","doi":"10.1145/3368235.3370929","DOIUrl":"https://doi.org/10.1145/3368235.3370929","url":null,"abstract":"Overlay network virtualization provides virtual networks over IP networks. Building virtual networks over IP networks achieves high mobility to realize a multi-tenant network. Although the encapsulation protocol, such as VXLAN, is widely used to implement overlay network virtualization, the encapsulation process degrades throughput. Therefore, we propose to combine NoEncap and Hoverboard model to improve throughput in commodity devices. NoEncap achieves performance enhancement of overlay network virtualization for commodity devices using non-encapsulation but increases the control plane overhead. The Hoverboard model reduces the control plane overhead. This paper presents the evaluation of the throughput enhancement by the proposed method.","PeriodicalId":166357,"journal":{"name":"Proceedings of the 12th IEEE/ACM International Conference on Utility and Cloud Computing Companion","volume":"17 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-12-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131275662","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Most organisations follow an open-source strategy when developing software products like cloud platforms. However, integrating various software pieces in a wrong way can lead to generating platforms that do not feature suitable non-functional properties and easily scale to handle an increase in customer base. Such platforms would thus not properly face the fierce competition in the cloud world. This paper proposes a novel methodology for selecting the best possible integration method and strategy for cloud platforms which enables addressing the previous challenges and producing platforms that can go beyond current competition. This methodology relies on investigating the current open-source platform components to be integrated and selecting the right integration method and strategy that leads to the best possible integration result by still keeping the integration effort to the minimum possible. This paper demonstrates its application on the MELODIC multi-cloud management platform, currently uptaken in the cloud market.
{"title":"Towards an Integration Methodology for Multi-Cloud Application Management Platforms","authors":"K. Kritikos, Pawel Skrzypek, M. Rózanska","doi":"10.1145/3368235.3368833","DOIUrl":"https://doi.org/10.1145/3368235.3368833","url":null,"abstract":"Most organisations follow an open-source strategy when developing software products like cloud platforms. However, integrating various software pieces in a wrong way can lead to generating platforms that do not feature suitable non-functional properties and easily scale to handle an increase in customer base. Such platforms would thus not properly face the fierce competition in the cloud world. This paper proposes a novel methodology for selecting the best possible integration method and strategy for cloud platforms which enables addressing the previous challenges and producing platforms that can go beyond current competition. This methodology relies on investigating the current open-source platform components to be integrated and selecting the right integration method and strategy that leads to the best possible integration result by still keeping the integration effort to the minimum possible. This paper demonstrates its application on the MELODIC multi-cloud management platform, currently uptaken in the cloud market.","PeriodicalId":166357,"journal":{"name":"Proceedings of the 12th IEEE/ACM International Conference on Utility and Cloud Computing Companion","volume":"65 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-12-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123187405","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
M. Mirzaei, Sanjeev Vellore Ranganathan, N. Kearns, David Airehrour, Mitra Etemaddar
Advances in machine learning and big data are leading to the employment of data analytics management systems in many organisations and industries. Generally, large organisations are able to put more resources towards this deployment, and the software used is more suited to larger entities. However, increasing business pressures are now driving small and medium-sized enterprises (SMEs) to deploy data analytics. This study examines the challenges faced by small retail businesses in New Zealand when deploying a self-service business intelligence (BI) system. The study applied qualitative methods including in-depth interviews with managers and staff of four retail stores, and observed that while respondents consistently agreed on the benefits and drawbacks of the new system, their levels of engagement with the system were uneven. It was discovered that users' real-time responses to dilemmas arising from the novelty of the new analytics management system compared to the existing system, significantly influenced the smooth adoption and implementation of the system. Analysis of a representative dilemma using the Theory of Constraints (TOC) and Evaporating Cloud (EC) method, leads to a possible resolution of the dilemma through TOC injections.
{"title":"Investigating Challenges to SME Deployment of Operational Business Intelligence: A Case Study in the New Zealand Retail Sector","authors":"M. Mirzaei, Sanjeev Vellore Ranganathan, N. Kearns, David Airehrour, Mitra Etemaddar","doi":"10.1145/3368235.3369371","DOIUrl":"https://doi.org/10.1145/3368235.3369371","url":null,"abstract":"Advances in machine learning and big data are leading to the employment of data analytics management systems in many organisations and industries. Generally, large organisations are able to put more resources towards this deployment, and the software used is more suited to larger entities. However, increasing business pressures are now driving small and medium-sized enterprises (SMEs) to deploy data analytics. This study examines the challenges faced by small retail businesses in New Zealand when deploying a self-service business intelligence (BI) system. The study applied qualitative methods including in-depth interviews with managers and staff of four retail stores, and observed that while respondents consistently agreed on the benefits and drawbacks of the new system, their levels of engagement with the system were uneven. It was discovered that users' real-time responses to dilemmas arising from the novelty of the new analytics management system compared to the existing system, significantly influenced the smooth adoption and implementation of the system. Analysis of a representative dilemma using the Theory of Constraints (TOC) and Evaporating Cloud (EC) method, leads to a possible resolution of the dilemma through TOC injections.","PeriodicalId":166357,"journal":{"name":"Proceedings of the 12th IEEE/ACM International Conference on Utility and Cloud Computing Companion","volume":"117 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-12-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124526866","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Daniel Ladouceur, Bimmy Pujari, Edward Gleeck, Joel Coffman
Mutual auditability offers visibility between a cloud service provider (CSP) and cloud service customer (CSC), informing both of the risks posed by their association with the other. In this work, we develop and experiment with two systems designed to enable such auditability: a specialized network-based intrusion detection system (NIDS) implementation, traditional-based intrusion system (TBIS), that gives CSPs insight into the malicious activity by clients' virtual machines (VMs) without undermining the CSC's privacy, and a complementary system, hypervisor-based intrusion system (HBIS), that provides visibility into malicious activities of co-resident CSCs by detecting side channel attacks. In order to ensure that our design does not introduce new vulnerabilities into the cloud environment, we examine the potential of using these auditing tools as attack vectors themselves and potential mitigations if such vulnerabilities are found.
{"title":"Techniques for Mutual Auditability in a Cloud Environment","authors":"Daniel Ladouceur, Bimmy Pujari, Edward Gleeck, Joel Coffman","doi":"10.1145/3368235.3368875","DOIUrl":"https://doi.org/10.1145/3368235.3368875","url":null,"abstract":"Mutual auditability offers visibility between a cloud service provider (CSP) and cloud service customer (CSC), informing both of the risks posed by their association with the other. In this work, we develop and experiment with two systems designed to enable such auditability: a specialized network-based intrusion detection system (NIDS) implementation, traditional-based intrusion system (TBIS), that gives CSPs insight into the malicious activity by clients' virtual machines (VMs) without undermining the CSC's privacy, and a complementary system, hypervisor-based intrusion system (HBIS), that provides visibility into malicious activities of co-resident CSCs by detecting side channel attacks. In order to ensure that our design does not introduce new vulnerabilities into the cloud environment, we examine the potential of using these auditing tools as attack vectors themselves and potential mitigations if such vulnerabilities are found.","PeriodicalId":166357,"journal":{"name":"Proceedings of the 12th IEEE/ACM International Conference on Utility and Cloud Computing Companion","volume":"53 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-12-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"117266952","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Taiyang Guo, R. Bahsoon, Tao-An Chen, Abdessalam Elhabbash, F. Samreen, Yehia El-khatib
Managing cloud spend and qualities when selecting cloud instances is cited as one of the timely research challenges in cloud computing. Cloud service consumers are often confronted by too many options and selection is challenging. This is because instance provision can be difficult to comprehend for an average technical user and tactics of cloud provider are far from being transparent biasing the selection. This paper proposes a novel cloud instance selection framework for finding the optimal IaaS purchase strategy for a VARD application in Amazon EC2. Analytical Hierarchy Process (AHP) and parallel K-Means Clustering algorithm are used and combined in Cloud Instance Selection environments. It allows cloud users to get the recommendation about cloud instance types and job submission periods based on requirements such as CPU, RAM, and resource utilisation. The system leverages AHP to select cloud instance type. Besides, AHP results are used by the parallel K-Means clustering model to find the best execution time for a given day according to the user's requirements. Finally, we provide an example to demonstrate the applicability of the approach. Experiments indicate that our approach achieves better results than ad-hoc and cost-driven approaches.
{"title":"Cloud Instance Selection Using Parallel K-Means and AHP","authors":"Taiyang Guo, R. Bahsoon, Tao-An Chen, Abdessalam Elhabbash, F. Samreen, Yehia El-khatib","doi":"10.1145/3368235.3368845","DOIUrl":"https://doi.org/10.1145/3368235.3368845","url":null,"abstract":"Managing cloud spend and qualities when selecting cloud instances is cited as one of the timely research challenges in cloud computing. Cloud service consumers are often confronted by too many options and selection is challenging. This is because instance provision can be difficult to comprehend for an average technical user and tactics of cloud provider are far from being transparent biasing the selection. This paper proposes a novel cloud instance selection framework for finding the optimal IaaS purchase strategy for a VARD application in Amazon EC2. Analytical Hierarchy Process (AHP) and parallel K-Means Clustering algorithm are used and combined in Cloud Instance Selection environments. It allows cloud users to get the recommendation about cloud instance types and job submission periods based on requirements such as CPU, RAM, and resource utilisation. The system leverages AHP to select cloud instance type. Besides, AHP results are used by the parallel K-Means clustering model to find the best execution time for a given day according to the user's requirements. Finally, we provide an example to demonstrate the applicability of the approach. Experiments indicate that our approach achieves better results than ad-hoc and cost-driven approaches.","PeriodicalId":166357,"journal":{"name":"Proceedings of the 12th IEEE/ACM International Conference on Utility and Cloud Computing Companion","volume":"86 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-12-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116083218","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Fog computing consists of modest capability fog nodes located close to the data generation sources. These nodes are ideal for executing small interactive tasks with a low latency requirement. Tasks that are larger and more latency tolerant may be executed at the cloud data center. A popular use case for fog computing is smart vehicles, which consist of numerous sensors and actuators that automate various tasks, such as traffic monitoring, braking and entertainment. We propose a fog computing based framework for pushing over the air (OTA) updates to smart vehicles. This allows the car manufacturers to push updates directly to the vehicles, without requiring a visit to the dealership. To this end, we propose a Software Update (SU ) algorithm that pushes OTA updates directly from fog nodes to the vehicles. Further, we propose a Mobility Management (MM) algorithm that takes into account the mobility of smart vehicles. In order to reduce the number of handovers, an ILP formulation has been proposed. Our experiments evaluate the impact of software size, vehicle mobility, number of vehicles and data transmission rate on OTA update performance. The experimental results using dynamic fog nodes offer an improvement of approximately 37% in OTA update time compared to updates using cloud data centers.
{"title":"Faster OTA Updates in Smart Vehicles using Fog Computing","authors":"Kaneez Fizza, Nitin Auluck, Akramul Azim, Md. Al Maruf, Anil Singh","doi":"10.1145/3368235.3368842","DOIUrl":"https://doi.org/10.1145/3368235.3368842","url":null,"abstract":"Fog computing consists of modest capability fog nodes located close to the data generation sources. These nodes are ideal for executing small interactive tasks with a low latency requirement. Tasks that are larger and more latency tolerant may be executed at the cloud data center. A popular use case for fog computing is smart vehicles, which consist of numerous sensors and actuators that automate various tasks, such as traffic monitoring, braking and entertainment. We propose a fog computing based framework for pushing over the air (OTA) updates to smart vehicles. This allows the car manufacturers to push updates directly to the vehicles, without requiring a visit to the dealership. To this end, we propose a Software Update (SU ) algorithm that pushes OTA updates directly from fog nodes to the vehicles. Further, we propose a Mobility Management (MM) algorithm that takes into account the mobility of smart vehicles. In order to reduce the number of handovers, an ILP formulation has been proposed. Our experiments evaluate the impact of software size, vehicle mobility, number of vehicles and data transmission rate on OTA update performance. The experimental results using dynamic fog nodes offer an improvement of approximately 37% in OTA update time compared to updates using cloud data centers.","PeriodicalId":166357,"journal":{"name":"Proceedings of the 12th IEEE/ACM International Conference on Utility and Cloud Computing Companion","volume":"15 5","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-12-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"120822552","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：