{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,8,2]],"date-time":"2026-08-02T10:26:18Z","timestamp":1785666378986,"version":"3.56.0"},"reference-count":57,"publisher":"MDPI AG","issue":"1","license":[{"start":{"date-parts":[[2020,12,29]],"date-time":"2020-12-29T00:00:00Z","timestamp":1609200000000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Future Internet"],"abstract":"<jats:p>The current approaches for energy consumption optimisation in buildings are mainly reactive or focus on scheduling of daily\/weekly operation modes in heating. Machine Learning (ML)-based advanced control methods have been demonstrated to improve energy efficiency when compared to these traditional methods. However, placing of ML-based models close to the buildings is not straightforward. Firstly, edge-devices typically have lower capabilities in terms of processing power, memory, and storage, which may limit execution of ML-based inference at the edge. Secondly, associated building information should be kept private. Thirdly, network access may be limited for serving a large number of edge devices. The contribution of this paper is an architecture, which enables training of ML-based models for energy consumption prediction in private cloud domain, and transfer of the models to edge nodes for prediction in Kubernetes environment. Additionally, predictors at the edge nodes can be automatically updated without interrupting operation. Performance results with sensor-based devices (Raspberry Pi 4 and Jetson Nano) indicated that a satisfactory prediction latency (~7\u20139 s) can be achieved within the research context. However, model switching led to an increase in prediction latency (~9\u201313 s). Partial evaluation of a Reference Architecture for edge computing systems, which was used as a starting point for architecture design, may be considered as an additional contribution of the paper.<\/jats:p>","DOI":"10.3390\/fi13010005","type":"journal-article","created":{"date-parts":[[2020,12,29]],"date-time":"2020-12-29T19:55:25Z","timestamp":1609271725000},"page":"5","update-policy":"https:\/\/doi.org\/10.3390\/mdpi_crossmark_policy","source":"Crossref","is-referenced-by-count":15,"title":["Architecture for Enabling Edge Inference via Model Transfer from Cloud Domain in a Kubernetes Environment"],"prefix":"10.3390","volume":"13","author":[{"given":"Pekka","family":"P\u00e4\u00e4kk\u00f6nen","sequence":"first","affiliation":[{"name":"VTT Technical Research Centre of Finland, 90571 Oulu, Finland"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Daniel","family":"Pakkala","sequence":"additional","affiliation":[{"name":"VTT Technical Research Centre of Finland, 90571 Oulu, Finland"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-0917-2242","authenticated-orcid":false,"given":"Jussi","family":"Kiljander","sequence":"additional","affiliation":[{"name":"VTT Technical Research Centre of Finland, 90571 Oulu, Finland"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Roope","family":"Sarala","sequence":"additional","affiliation":[{"name":"VTT Technical Research Centre of Finland, 90571 Oulu, Finland"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1968","published-online":{"date-parts":[[2020,12,29]]},"reference":[{"key":"ref_1","unstructured":"Annual Energy Outlook Early Release (2020, November 18). Energy Information Administration (EIA), Available online: https:\/\/www.eia.gov\/outlooks\/aeo\/."},{"key":"ref_2","unstructured":"Rousselot, M. (2020, November 18). ODYSSEE-MURE Policy Brief: Energy Efficiency Trends in Buildings. Available online: https:\/\/www.odyssee-mure.eu\/publications\/policy-brief\/buildings-energy-efficiency-trends.html."},{"key":"ref_3","unstructured":"(2020, November 18). Heating Cooling. Available online: https:\/\/ec.europa.eu\/energy\/topics\/energy-efficiency\/heating-and-cooling_en."},{"key":"ref_4","doi-asserted-by":"crossref","first-page":"8300","DOI":"10.3390\/en8088300","article-title":"Learning agent for a heat-pump thermostat with a set-back strategy using model-free reinforcement learning","volume":"8","author":"Ruelens","year":"2015","journal-title":"Energies"},{"key":"ref_5","doi-asserted-by":"crossref","unstructured":"Barrett, E., and Linder, S. (2015, January 7\u201311). Autonomous hvac control, a reinforcement learning approach. Proceedings of the Joint European Conference on Machine Learning and Knowledge Discovery in Databases, Porto, Portugal.","DOI":"10.1007\/978-3-319-23461-8_1"},{"key":"ref_6","doi-asserted-by":"crossref","first-page":"6158","DOI":"10.1016\/j.egypro.2019.01.494","article-title":"Advanced building control via deep reinforcement learning","volume":"158","author":"Jia","year":"2019","journal-title":"Energy Procedia"},{"key":"ref_7","unstructured":"Zhang, Z., Chong, A., Pan, Y., Zhang, C., Lu, S., and Lan, K.P. (2018, January 26\u201328). A Deep Reinforcement Learning Approach to Using Whole Building Energy Model for HVAC Optimal Control. Proceedings of the Building Performance Modeling Conference and SIMBuild, Chicago, IL, USA."},{"key":"ref_8","doi-asserted-by":"crossref","first-page":"96","DOI":"10.1016\/j.enbuild.2017.02.012","article-title":"Artificial neural network (ANN) based model predictive control (MPC) and optimization of HVAC systems: A state of the art review and case study of a residential HVAC system","volume":"141","author":"Afram","year":"2017","journal-title":"Energy Build."},{"key":"ref_9","unstructured":"Jain, A., Smarra, F., Reticcioli, E., D\u2019Innocenzo, A., and Morari, M. (2020, January 11\u201312). NeurOpt: Neural network based optimization for building energy management and climate control. Proceedings of the 2nd Annual Conference on Learning for Dynamics and Control, Berkeley, CA, USA."},{"key":"ref_10","doi-asserted-by":"crossref","first-page":"30","DOI":"10.1016\/j.rser.2016.03.047","article-title":"Smart home energy management systems: Concept, configurations, and scheduling strategies","volume":"61","author":"Zhou","year":"2016","journal-title":"Renew. Sustain. Energy Rev."},{"key":"ref_11","doi-asserted-by":"crossref","unstructured":"Hadidi, R., Cao, J., Xie, Y., Asgari, B., Krishna, T., and Kim, H. (2019, January 3\u20135). Characterizing the Deployment of Deep Neural Networks on Commercial Edge Devices. Proceedings of the IEEE International Symposium on Workload Characterization (IISWC), Orlando, FL, USA.","DOI":"10.1109\/IISWC47752.2019.9041955"},{"key":"ref_12","doi-asserted-by":"crossref","first-page":"1655","DOI":"10.1109\/JPROC.2019.2921977","article-title":"Deep learning with edge computing: A review","volume":"107","author":"Chen","year":"2019","journal-title":"Proc. IEEE"},{"key":"ref_13","doi-asserted-by":"crossref","unstructured":"Pakkala, D., and Spohrer, J. (2019, January 8\u201311). Digital Service: Technological Agency in Service Systems. Proceedings of the Hawaii International Conference on System Sciences, Honolulu, HI, USA.","DOI":"10.24251\/HICSS.2019.229"},{"key":"ref_14","doi-asserted-by":"crossref","first-page":"44","DOI":"10.1145\/3331138","article-title":"Research for practice: The DevOps phenomenon","volume":"62","author":"Wiedemann","year":"2019","journal-title":"Comm. ACM"},{"key":"ref_15","doi-asserted-by":"crossref","unstructured":"Hummer, W., Muthusamy, V., Rausch, T., Dube, P., El Maghraoui, K., Murthi, A., and Oum, P. (2019, January 24\u201327). ModelOps: Cloud-based lifecycle management for reliable and trusted AI. Proceedings of the IEEE International Conference on Cloud Engineering (IC2E), Prague, Czech Republic.","DOI":"10.1109\/IC2E.2019.00025"},{"key":"ref_16","doi-asserted-by":"crossref","first-page":"278","DOI":"10.1016\/j.future.2019.04.016","article-title":"A review of edge computing reference architectures and a new global edge proposal","volume":"99","author":"Alonso","year":"2019","journal-title":"Future Gener. Comput. Syst."},{"key":"ref_17","first-page":"1","article-title":"Extending reference architecture of big data systems towards machine learning in edge computing environments","volume":"25","author":"Pakkala","year":"2020","journal-title":"J. Big Data"},{"key":"ref_18","unstructured":"(2020, November 18). Rancher. Available online: https:\/\/rancher.com\/."},{"key":"ref_19","unstructured":"(2020, November 18). k3s-Lightweight Kubernetes. Available online: https:\/\/rancher.com\/docs\/k3s\/latest\/en\/."},{"key":"ref_20","doi-asserted-by":"crossref","first-page":"444","DOI":"10.1016\/j.future.2019.06.030","article-title":"A short-term energy prediction system based on edge computing for smart city","volume":"101","author":"Luo","year":"2019","journal-title":"Future Gener. Comput. Syst."},{"key":"ref_21","doi-asserted-by":"crossref","unstructured":"Li, K., and Gui, N. (2020). CMS: A Continuous Machine-Learning and Serving Platform for Industrial Big Data. Future Internet, 12.","DOI":"10.3390\/fi12060102"},{"key":"ref_22","doi-asserted-by":"crossref","first-page":"45","DOI":"10.2753\/MIS0742-1222240302","article-title":"A Design Science Research Methodology for Information Systems Research","volume":"24","author":"Peffers","year":"2007","journal-title":"J. Manag. Inf. Syst."},{"key":"ref_23","doi-asserted-by":"crossref","unstructured":"Salmi, T., Kiljander, J., and Pakkala, D. (2020). Stacked Boosters Network Architecture for Short-Term Load Forecasting in Buildings. Energies, 13.","DOI":"10.3390\/en13092370"},{"key":"ref_24","unstructured":"(2020, November 18). ISO\/IEC JTC1\/SC 42 Committee. Available online: https:\/\/www.iso.org\/committee\/6794475.html."},{"key":"ref_25","unstructured":"Big Data Value Association (2020, November 18). BVA SRIA\u2014European Big Data Value Strategic Research and Innovation Agenda. Available online: http:\/\/bdva.eu\/sites\/default\/files\/BDVA_SRIA_v4_Ed1.1.pdf."},{"key":"ref_26","unstructured":"Chang, W.L., Boyd, D., and Levin, O. (2020, November 18). NIST Big Data Interoperability Framework: Volume 6, Reference Architecure. NIST Big Data Program, Available online: https:\/\/www.nist.gov\/publications\/nist-big-data-interoperability-framework-volume-6-reference-architecture."},{"key":"ref_27","unstructured":"Lin, S., and Simmon, E. (2019). The Industrial Internet of Things Volume G1: Reference Architecture, Industrial Internet Consortium."},{"key":"ref_28","unstructured":"(2020, November 18). ArchiMate 3.1 Specification. The Open Group. Available online: https:\/\/pubs.opengroup.org\/architecture\/archimate3-doc\/."},{"key":"ref_29","doi-asserted-by":"crossref","unstructured":"Dang, Y., Lin, Q., and Huang, P. (2019, January 25\u201331). AIOps: Real-World challenges and research innovations. Proceedings of the IEEE\/ACM 41st International Conference on Software Engineering: Companion Proceedings (ICSE-Companion), Montreal, QC, Canada.","DOI":"10.1109\/ICSE-Companion.2019.00023"},{"key":"ref_30","doi-asserted-by":"crossref","unstructured":"Galster, M., and Avgeriou, P. (2011, January 20\u201324). Empirically-grounded reference architectures: A proposal. Proceedings of the Joint ACM SIGSOFT Conference on Quality of Software Architectures and ACM SIGSOFT Conference on Quality of Software Architectures and ACM SIGSOFT Symposium on Architecting Critical Systems, Boulder, CO, USA.","DOI":"10.1145\/2000259.2000285"},{"key":"ref_31","unstructured":"Hardy, C., Merrer, E.L., and Sericola, B. (November, January 30). Distributed deep learning on edge-devices: Feasibility via adaptive compression. Proceedings of the IEEE 16th International Symposium on Network Computing and Applications (NCA), Cambridge, MA, USA."},{"key":"ref_32","doi-asserted-by":"crossref","unstructured":"Jeong, H., Jeong, I., Lee, H., and Moon, S. (2019, January 2\u20135). Computation offloading for machine learning web apps in the edge server environment. Proceedings of the IEEE 38th International Conference on Distributed Computing Systems, Vienna, Austria.","DOI":"10.1109\/ICDCS.2018.00154"},{"key":"ref_33","unstructured":"Zhou, L., Wen, H., Teodorescu, R., and Du, D.H.C. (2019, January 9). Distributing deep neural networks with containerized partitions at the edge. Proceedings of the 2nd Usenix Workshop on Hot Topics in Edge Computing, Renton, WA, USA."},{"key":"ref_34","doi-asserted-by":"crossref","unstructured":"Mehta, R., and Shorey, R. (2020, January 7\u201311). DeepSplit: Dynamic Splitting of Collaborative Edge-Cloud Convolutional Neural Networks. Proceedings of the 12th International Conference on Communication Systems  Networks (COMSNETS), Bengaluru, India.","DOI":"10.1109\/COMSNETS48256.2020.9027432"},{"key":"ref_35","unstructured":"Hadidi, R., Asgari, B., Cao, J., Bae, Y., Kim, H., Ryoo, M.S., and Kim, H. (2020, November 18). Edge-Tailored Perception: Fast Inferencing in-the-Edge with Efficient Model Distribution. Available online: https:\/\/deepai.org\/publication\/edge-tailored-perception-fast-inferencing-in-the-edge-with-efficient-model-distribution."},{"key":"ref_36","doi-asserted-by":"crossref","first-page":"4950","DOI":"10.1109\/JIOT.2020.2972000","article-title":"Toward Collaborative Inferencing of Deep Neural Networks on Internet-of-Things Devices","volume":"7","author":"Hadidi","year":"2020","journal-title":"IEEE Internet Things"},{"key":"ref_37","doi-asserted-by":"crossref","unstructured":"Ran, X., Chen, H., Chu, X., Liu, Z., and Chen, J. (2018, January 16\u201319). DeepDecision: A Mobile deep learning framework for edge video analytics. Proceedings of the IEEE Conference on Computer Communications, Honolulu, HI, USA.","DOI":"10.1109\/INFOCOM.2018.8485905"},{"key":"ref_38","first-page":"1","article-title":"Docker: Lightweight Linux Containers for Consistent Development and Deployment","volume":"239","author":"Merkel","year":"2014","journal-title":"Linux J."},{"key":"ref_39","doi-asserted-by":"crossref","first-page":"50","DOI":"10.1145\/2890784","article-title":"Borg, Omega, and Kubernetes","volume":"59","author":"Burns","year":"2016","journal-title":"Commun. ACM"},{"key":"ref_40","unstructured":"(2020, November 18). Helm. The Package Manager for Kubernetes. Available online: https:\/\/helm.sh\/."},{"key":"ref_41","doi-asserted-by":"crossref","unstructured":"Fathoni, H., Yang, C., Chang, C., and Huang, C. (2019, January 16\u201320). Performance Comparison of Lightweight Kubernetes in Edge Devices. Proceedings of the I-SPAN: International Symposium on Pervasive Systems, Algorithms and Networks, Naples, Italy.","DOI":"10.1007\/978-3-030-30143-9_25"},{"key":"ref_42","doi-asserted-by":"crossref","unstructured":"Goethals, T., De Turck, F., and Volckaert, B. (2019, January 18\u201321). FLEDGE: Kubernetes Compatible Container Orchestration on Low-Resource Edge Devices. Proceedings of the IOV: International Conference on Internet of Vehicles, Kaohsiung, Taiwan.","DOI":"10.1007\/978-3-030-38651-1_16"},{"key":"ref_43","doi-asserted-by":"crossref","unstructured":"Le Minh, K., Le, K., and Le-Trung, Q. (2020, January 28\u201329). DLASE: A light-weight framework supporting Deep Learning for Edge Devices. Proceedings of the 4th International Conference on Recent Advances in Signal Processing, Telecommunications  Computing (SigTelCom), Hanoi, Vietnam.","DOI":"10.1109\/SigTelCom49868.2020.9199058"},{"key":"ref_44","doi-asserted-by":"crossref","unstructured":"Lee, S.H., Lee, T., Kim, S., and Park, S. (2019, January 10\u201313). Energy Consumption Prediction System Based on Deep Learning with Edge Computing. Proceedings of the 2nd International Conference on Electronics Technology, Chengdu, China.","DOI":"10.1109\/ELTECH.2019.8839589"},{"key":"ref_45","doi-asserted-by":"crossref","unstructured":"Sonnenberg, C., and vom Brocke, J. (2012, January 14\u201315). Evaluations in the Science of the Artificial\u2013Reconsidering the Build-Evaluate Pattern in Design Science Research. Proceedings of the International Conference on Design Science Research in Information Systems and Technology, Las Vegas, NV, USA.","DOI":"10.1007\/978-3-642-29863-9_28"},{"key":"ref_46","doi-asserted-by":"crossref","first-page":"3125","DOI":"10.1109\/TSG.2018.2818167","article-title":"Review of Smart Meter Data Analytics: Applications, Methodologies, and Challenges","volume":"10","author":"Wang","year":"2019","journal-title":"IEEE Trans. Smart Grid"},{"key":"ref_47","unstructured":"(2020, November 18). Nvidia Tesla P100. Available online: https:\/\/www.nvidia.com\/en-us\/data-center\/tesla-p100\/."},{"key":"ref_48","unstructured":"(2020, November 18). Persistent Volumes. Available online: https:\/\/kubernetes.io\/docs\/concepts\/storage\/persistent-volumes\/."},{"key":"ref_49","unstructured":"(2020, November 18). Docker Buildx. Available online: https:\/\/docs.docker.com\/buildx\/working-with-buildx\/."},{"key":"ref_50","unstructured":"(2020, November 18). NVIDIA L4T TensorFlow. Available online: https:\/\/ngc.nvidia.com\/catalog\/containers\/nvidia:l4t-tensorflow."},{"key":"ref_51","unstructured":"(2020, November 18). hellozcb\/tensorflow-arm64 at Docker Hub. Available online: https:\/\/hub.docker.com\/r\/hellozcb\/tensorflow-arm64."},{"key":"ref_52","doi-asserted-by":"crossref","first-page":"166","DOI":"10.1016\/j.bdr.2015.01.001","article-title":"Reference Architecture and Classification of Technologies, Products and Services for Big Data Systems","volume":"2","author":"Pakkala","year":"2015","journal-title":"Big Data Res."},{"key":"ref_53","doi-asserted-by":"crossref","unstructured":"Lipcak, P., Macak, M., and Rossi, B. (2019, January 1\u20134). Big Data Platform for Smart Grids Power Consumption Anomaly Detection. Proceedings of the Federated Conference on Computer Science and Information Systems, Leipzig, Germany.","DOI":"10.15439\/2019F210"},{"key":"ref_54","doi-asserted-by":"crossref","first-page":"37","DOI":"10.1016\/j.infsof.2017.03.011","article-title":"Benefits and drawbacks of software reference architectures: A case study","volume":"88","author":"Ayala","year":"2017","journal-title":"Inf. Softw. Technol."},{"key":"ref_55","unstructured":"(2020, November 18). TensorFlow Lite. Available online: https:\/\/www.tensorflow.org\/lite."},{"key":"ref_56","unstructured":"(2020, November 18). Docker Registry. Available online: https:\/\/docs.docker.com\/registry\/."},{"key":"ref_57","unstructured":"(2020, November 18). k3s issue (#1278) in GitHub. Available online: https:\/\/github.com\/rancher\/k3s\/issues\/1278."}],"container-title":["Future Internet"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.mdpi.com\/1999-5903\/13\/1\/5\/pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,10,11]],"date-time":"2025-10-11T10:47:37Z","timestamp":1760179657000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.mdpi.com\/1999-5903\/13\/1\/5"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2020,12,29]]},"references-count":57,"journal-issue":{"issue":"1","published-online":{"date-parts":[[2021,1]]}},"alternative-id":["fi13010005"],"URL":"https:\/\/doi.org\/10.3390\/fi13010005","relation":{},"ISSN":["1999-5903"],"issn-type":[{"value":"1999-5903","type":"electronic"}],"subject":[],"published":{"date-parts":[[2020,12,29]]}}}