The serverless computing paradigm has transformed cloud service deployment by enabling automatic scaling of resources in response to varying demand. Building on this, stateful serverless computing introduces critical capabilities for data management, fault tolerance, and consistency, which are particularly relevant in the context of distributed deployments, notably in edge computing environments. In this work, we explore the feasibility of stateful serverless computing in resource-limited edge environments through an empirical study utilizing a multi-view object tracking application. Our results show that while these systems perform well in cloud environments, their effectiveness is severely affected at the edge due to state, application, and resource management solutions optimized for cloud environments. Existing solutions are most detrimental to applications with intermittent workloads, as typical combinations of concurrency handling and resource reservation can lead to minutes of unstable system behavior due to cold starts. Our results highlight the need for a tailored approach in stateful serverless systems for edge computing scenarios. 

Cloud formations often obscure optical satellite-based monitoring of the Earth’s surface, thus limiting Earth observation (EO) activities such as land cover mapping, ocean color analysis, and cropland monitoring. The integration of machine learning (ML) methods within the remote sensing domain has significantly improved performance for a wide range of EO tasks, including cloud detection and filtering, but there is still much room for improvement. A key bottleneck is that ML methods typically depend on large amounts of annotated data for training, which are often difficult to come by in EO contexts. This is especially true when it comes to cloud optical thickness (COT) estimation. A reliable estimation of COT enables more fine-grained and application-dependent control compared to using pre-specified cloud categories, as is common practice. To alleviate the COT data scarcity problem, in this work, we propose a novel synthetic dataset for COT estimation, which we subsequently leverage for obtaining reliable and versatile cloud masks on real data. In our dataset, top-of-atmosphere radiances have been simulated for 12 of the spectral bands of the Multispectral Imagery (MSI) sensor onboard Sentinel-2 platforms. These data points have been simulated under consideration of different cloud types, COTs, and ground surface and atmospheric profiles. Extensive experimentation of training several ML models to predict COT from the measured reflectivity of the spectral bands demonstrates the usefulness of our proposed dataset. In particular, by thresholding COT estimates from our ML models, we show on two satellite image datasets (one that is publicly available, and one which we have collected and annotated) that reliable cloud masks can be obtained. The synthetic data, the newly collected real dataset, code and models have been made publicly available. 

Emerging applications in IoT (Internet of Things) and edge computing/learning have sparked massive renewed interest in developing distributed versions of existing (centralized) iterative algorithms often used for optimization or machine learning purposes. While existing work in the literature exhibit similarities, for the tasks of both algorithm design and theoretical analysis, there is still no unified method or framework for accomplishing these tasks. This paper develops such a general framework, for distributing the execution of (centralized) iterative algorithms over networks in which the required information or data is partitioned between the nodes in the network. This paper furthermore shows that the distributed iterative algorithm, which results from the proposed framework, retains the convergence properties (rate) of the original (centralized) iterative algorithm. In addition, this paper applies the proposed general framework to several interesting example applications, obtaining results comparable to the state of the art for each such example, while greatly simplifying and generalizing their convergence analysis. These example applications reveal new results for distributed proximal versions of gradient descent, the heavy-ball method, and Newton's method. For example, these results show that the dependence on the condition number for the convergence rate of this distributed Heavy ball method is at least as good as for centralized gradient descent. Author

Current district heating networks are undergoing a sustainable transition towards the 4th and 5th generation of district heating networks, characterized by the integration of different types of renewable energy sources (RES) and low operational temperatures, i.e., 55°C or lower. Due to the lower temperature difference between supply and return, it is necessary to develop novel methods to understand the loads accurately and provide operation scenarios to anticipate demand peaks and increase flexibility in the energy network, both for long- and short-term horizons. In this study, a hybrid machine-learning (ML) method is developed, combining a clustering pre-processing step with a multi-input artificial neural network (ANN) model to predict heat loads in buildings cluster-wise. Specifically, the impact of time-series data clustering, as a pre-processing step, on the performance of ML models was investigated. It was found that data clustering contributes effectively to the reduction of data training costs by limiting the training processes to representative clusters only instead of all datasets. Additionally, low-quality data, including outliers and large measurement gaps, are excluded from the training to enhance the overall prediction performance of the models.

Energy efficient control of energy systems in buildings is a widely recognized challenge due to the use of low temperature heating, renewable electricity sources, and the incorporation of thermal storage. Reinforcement Learning (RL) has been shown to be effective at minimizing the energy usage in buildings with maintained thermal comfort despite the high system complexity. However, RL has certain disadvantages that make it challenging to apply in engineering practices. In this review, we take a computer science approach to identifying three main categories of challenges of using RL for control of Building Energy Systems (BES). The three categories are the following: RL in single buildings, RL in building clusters, and multi-agent aspects. For each topic, we analyse the main challenges, and the state-of-the-art approaches to alleviate them. We also identify several future research directions on subjects such as sample efficiency, transfer learning, and the theoretical properties of RL in building energy systems. In conclusion, our review shows that the work on RL for BES control is still in its initial stages. Although significant progress has been made, more research is needed to realize the goal of RL-based control of BES at scale.

Use of the serverless paradigm in cloud application development is growing rapidly, primarily driven by its promise to free developers from the responsibility of provisioning, operating, and scaling the underlying infrastructure. However, modern cloud-edge infrastructures are characterized by large numbers of disparate providers, constrained resource devices, platform heterogeneity, infrastructural dynamicity, and the need to orchestrate geographically distributed nodes and devices over public networks. This presents significant management complexity that must be addressed if serverless technologies are to be used in production systems. This position paper introduces COGNIT, a major new European initiative aiming to integrate AI technology into cloud-edge management systems to create a Cognitive Cloud reference framework and associated tools for serverless computing at the edge. COGNIT aims to: 1) support an innovative new serverless paradigm for edge application management and enhanced digital sovereignty for users and developers; 2) enable on-demand deployment of large-scale, highly distributed and self-adaptive serverless environments using existing cloud resources; 3) optimize data placement according to changes in energy efficiency heuristics and application demands and behavior; 4) enable secure and trusted execution of serverless runtimes. We identify and discuss seven research challenges related to the integration of serverless technologies with multi-provider Edge infrastructures and present our vision for how these challenges can be solved. We introduce a high-level view of our reference architecture for serverless cloud-edge continuum systems, and detail four motivating real-world use cases that will be used for validation, drawing from domains within Smart Cities, Agriculture and Environment, Energy, and Cybersecurity. 

In this work, we develop machine learning methods to forecast the day-ahead heating energy demand of district heating (DH) end-users in hourly resolution, using existing metering data for DH end-users and weather data. The focus of the study is a detailed analysis of the accuracy levels of short-term load prediction methods. In particular, accuracy levels are quantified for Artificial Neural Network (ANN) models with variations in the input parameters. The importance of historical data is investigated – in particular the importance of including historical hourly heating loads as input to the forecasting model. Additionally, the impact of different lengths of the historical input data is studied. Our methods are evaluated and validated using metering data from a live use-case in a Scandinavian environment, collected from 20 DH-supplied nursing homes through the years of 2016 to 2019. This study demonstrates that, although there is a strong linear relationship between outdoor temperature and heating load, it is still important to include historical heating loads as an input for prediction of future heating loads. Furthermore, the results show that it is important to include historical data from at least the preceding 24 hours, but suggest diminishing returns of including data much further back than that. The resulting models demonstrate the practical feasibility of such prediction models in a live use-case.

In this work, we develop machine learning methods to forecast the day-ahead heating energy demand of district heating (DH) end-users in hourly resolution, using existing metering data for DH end-users and weather data. The focus of the study is a detailed analysis of the accuracy levels of short-term load prediction methods. In particular, accuracy levels are quantified for Artificial Neural Network (ANN) models with variations in the input parameters. The importance of historical data is investigated – in particular the importance of including historical hourly heating loads as input to the forecasting model. Additionally, the impact of different lengths of the historical input data is studied. Our methods are evaluated and validated using metering data from a live use-case in a Scandinavian environment, collected from 20 DH-supplied nursing homes through the years of 2016 to 2019. This study demonstrates that, although there is a strong linear relationship between outdoor temperature and heating load, it is still important to include historical heating loads as an input for prediction of future heating loads. Furthermore, the results show that it is important to include historical data from at least the preceding 24 hours, but suggest diminishing returns of including data much further back than that. The resulting models demonstrate the practical feasibility of such prediction models in a live use-case.

In the face of green energy initiatives and progressively increasing shares of more energy-efficient buildings, there is a pressing need to transform district heating towards low-temperature district heating. The substantially lowered supply temperature of low-temperature district heating broadens the opportunities and challenges to integrate distributed renewable energy, which requires enhancement on intelligent heating load prediction. Meanwhile, to fulfill the temperature requirements for domestic hot water and space heating, separate energy conversion units on user-side, such as building-sized boosting heat pumps shall be implemented to upgrade the temperature level of the low-temperature district heating network. This study conducted hybrid heating load prediction methods with long-term and short-term prediction, and the main work consisted of four steps: (1) acquisition and processing of district heating data of 20 district heating supplied nursing homes in the Nordic climate (2016–2019); (2) long-term district heating load prediction through linear regression, energy signature curve in hourly resolution, providing an overall view and boundary conditions for the unit sizing; (3) short-term district heating load prediction through two Artificial Neural Network models, f72 and g120, with different prediction input parameters; (4) evaluation of the predicted load profiles based on the measured data. Although the three prediction models met the quality criteria, it was found that including the historical hourly heating loads as the input to the forecasting model enhanced the prediction quality, especially for the peak load and low-mild heating season. Furthermore, a possible application of the heating load profiles was proposed by integrating two building-sized heat pumps in low-temperature district heating, which may be a promising heat supply method in low-temperature district heating. © 2022 The Authors