Smart City Gnosys

Smart city article details

Title Serverless Data Pipelines For Iot Data Analytics: A Cloud Vendors Perspective And Solutions
ID_Doc 48501
Authors Poojara S.; Dehury C.K.; Jakovits P.; Srirama S.N.
Year 2022
Published Predictive Analytics in Cloud, Fog, and Edge Computing: Perspectives and Practices of Blockchain, IoT, and 5G
DOI http://dx.doi.org/10.1007/978-3-031-18034-7_7
Abstract Advancements in Internet and Telecommunications (ICT) accelerated the large-scale deployments of IoT applications including smart city, smart healthcare, and smart factory aiming for faster data processing. The latency and other challenges of cloud-centric IoT, driven edge computing-based data processing architectures. In precise, IoT data analytics requires control for dealing with the complete life cycle of data flow between data sources to sinks by building a set of data pipelines seamlessly deployed on the IoT computing continuum (Edge and Clouds). The execution of data analytic tasks is challenging in the IoT computing continuum due to heterogeneous hardware architecture at edge and cloud environments. This can be succeeded using Serverless or Function as a Service (FaaS) cloud computing model, wherein tasks are defined as virtual functions and seamlessly migrated and executed within the computing continuum. The serverless data pipelines constitutes a data source, a set of serverless functions that performs specific operations on the data and then ultimately a data sink. Similarly, AWS, Microsoft Azure and Google Cloud Providers have typical edge and cloud solutions for IoT data processing that includes the serverless entities as a part of the system. However, cost and latency vary according to cloud provider subscriptions and architecture components. So our proposed work outlines the cloud service provider-specific IoT services that are composed in designing the serverless data pipelines (SDP) for IoT data processing. Further, we investigate their performance in terms of latency, cost and cold/warm start time of serverless functions for IoT data oriented pipelines. Aligning to this, we propose AWS and Microsoft Azure based SDP architectures for predictive analytics in Industrial IoT environments and implement the real time Predictive Maintenance of Industrial Motor application. Accordingly, we measure the efficiency of AWS and Azure SDP architecture for various performance metrics such as cost, processing time and cold/warm start time for serverless invocations. © The Editor(s) (if applicable) and The Author(s), under exclusive license to Springer Nature Switzerland AG 2023.
Author Keywords Data pipelines; IoT; Predictive maintenance; Serverless


Similar Articles


Id Similarity Authors Title Published
30629 View0.872Asha A.; Rajeshkumar L.; Pandi V.S.; Shobana D.; Lakshmi Priya J.; Dayanidhy M.Implementing Cloud Computing With Internet Of Things (Iot) Technologies: Novel Approaches To Data Management And Service Delivery Innovation2024 Global Conference on Communications and Information Technologies, GCCIT 2024 (2024)
7612 View0.87Sabbioni A.; Bujari A.; Romeo S.; Foschini L.; Corradi A.An Architectural Approach For Heterogeneous Data Access In Serverless PlatformsProceedings - IEEE Global Communications Conference, GLOBECOM (2022)
1456 View0.867Asaad R.R.; Hani A.A.; Sallow A.B.; Abdulrahman S.M.; Ahmad H.B.; Subhi R.M.A Development Of Edge Computing Method In Integration With Iot System For Optimizing And To Produce Energy Efficiency System2024 4th International Conference on Advance Computing and Innovative Technologies in Engineering, ICACITE 2024 (2024)
8445 View0.865Mohan S.; Panda S.An Integrated Framework For Real-Time Analysis And Observability Of Wireless Sensor Data Using Aws Edge Service CapabilitiesProceedings of 2024 IEEE International Conference of Electron Devices Society Kolkata Chapter, EDKCON 2024 (2024)
13107 View0.862Krishnappa M.S.; Jayabalan D.; Harve B.M.; Jayaram V.; Bidkar D.M.; Veerapaneni P.K.; Gejjegondanahalli V.Y.Building The Future Of Iot: Cloud Platforms, Integration Challenges, And Emerging Applications2024 International Conference on Computer and Applications, ICCA 2024 (2024)
14631 View0.859Townend P.; Marti A.P.; De La Iglesia I.; Matskanis N.; Timoudas T.O.; Hallmann T.; Lalaguna A.; Swat K.; Renzi F.; Bochenski D.; Mancini M.; Bhuyan M.; Gonzalez-Hierro M.; Dupont S.; Kristiansson J.; Montero R.S.; Elmroth E.; Valdes I.; Massonet P.; Olsson D.; Llorente I.M.; Ostberg P.-O.; Abdou M.Cognit: Challenges And Vision For A Serverless And Multi-Provider Cognitive Cloud-Edge ContinuumProceedings - IEEE International Conference on Edge Computing, 2023-July (2023)
42873 View0.856Thakkar H.K.; Dehury C.K.; Sahoo P.K.; Veeravalli B.Predictive Analytics In Cloud, Fog, And Edge Computing: Perspectives And Practices Of Blockchain, Iot, And 5GPredictive Analytics in Cloud, Fog, and Edge Computing: Perspectives and Practices of Blockchain, IoT, and 5G (2022)
33904 View0.854Kilaru M.; Maheswari P.; Boddepalli E.; Venkataramana K.; Patel J.D.; Sharma M.K.Iot Services And Intelligence: Empowering The Internet Of Things With Real-Time Data Analytics And Decision-MakingTQCEBT 2024 - 2nd IEEE International Conference on Trends in Quantum Computing and Emerging Business Technologies 2024 (2024)
55982 View0.853Hasan M.M.; Sultana T.; Hossain M.D.; Mandal A.K.; Ngo T.-T.; Lee G.-W.; Huh E.-N.The Journey To Cloud As A Continuum: Opportunities, Challenges, And Research DirectionsICT Express (2025)
41780 View0.853Trieu Q.L.; Javadi B.; Basilakis J.; Toosi A.N.Performance Evaluation Of Serverless Edge Computing For Machine Learning ApplicationsProceedings - 2022 IEEE/ACM 15th International Conference on Utility and Cloud Computing, UCC 2022 (2022)