The increasing demand for sustainable and real-time environmental monitoring has highlighted the need for efficient energy utilization in solar-powered systems. This research introduces a convolutional transformer-based predictive model for solar energy forecasting, representing a significant advancement in leveraging machine learning for Internet of Things (IoT) and environmental monitoring applications. Unlike prior approaches that primarily relied on conventional methods or basic predictive models, this work integrates advanced techniques such as Variational Mode Decomposition (VMD) to preprocess data and enhance forecasting accuracy. These improvements directly enhance the entire operational framework, enabling not only optimized energy utilization but also reduced system downtime and overutilization. By focusing on forecasting, this research contribution advances the state-of-the-art in predictive modeling for energy management in autonomous monitoring systems. Accurate solar energy forecasts empower dynamic adjustments in system parameters, ensuring both operational efficiency and sustainability. Comparative analysis demonstrates the substantial gains achieved by the proposed approach, highlighting its potential to transform the feasibility and performance of autonomous environmental monitoring systems. Additionally, this work integrates advanced predictive models into solar-powered frameworks, contributing to sustainable and efficient environmental monitoring practices.

The Berkeley Packet Filter (BPF) is increasingly used for a variety of use cases including auditing, security, monitoring, networking, etc. However, BPF lacks improved and effective tools, which makes the integration of programs written for BPF quite hard in current automated testing solutions and continue integration pipelines. We present tBPF, a library for integration testing of BPF programs that allows automated testing of arbitrary programs in a kernel agnostic manner without superuser privileges. We show that our solution can be integrated in existing development workflows and pipelines to enable reproducible testing and auditing of BPF programs. In this article, we describe our approach and compare it against other approaches in relevant work and literature.

In light of the essential need for persistent, real-time monitoring of natural environments, our research introduces an innovative approach that integrates a predictive solar energy model in optimizing solar-powered environmental monitoring efficiency with energy constraints. We address the challenge of solar energy variability by forecasting solar energy availability based on the Random Forest model. Then, we integrate the prediction ability into the Integer Linear Program optimization framework of the monitoring efficiency subjected to energy constraints. Experimental results demonstrate our approach’s effectiveness compared to traditional methods, underscoring its potential for enhancing sensor networks’ sustainability and operational efficiency in natural environments.

The rapid advancement of 5G technology has significantly increased energy consumption, underscoring the need for advanced energy management solutions. Proactive energy management, which relies on accurate predictions of network load to enable timely adaptive actions, emerges as a key strategy in addressing this challenge. In this study, we introduce a refined approach to forecasting traffic load in 5G networks, emphasizing the integration of statistical and temporal feature engineering. This methodology is aimed at capturing the intricate spatial and temporal patterns inherent in network data, thereby enhancing prediction accuracy. Leveraging an existing dataset comprising measurements from 1,000 base stations, we enriched this dataset with a set of derived features that reflect both temporal dynamics and load characteristics. Utilizing this enriched dataset, we trained and validated a suite of predictive models. Our findings reveal a notable improvement in the accuracy of traffic load predictions, outperforming standard baseline models. This underscores the effectiveness of our feature engineering approach in refining the predictive capabilities of models, paving the way for more efficient and proactive energy management in 5G networks.

The energy consumption growth of the Information and Communication Technology (ICT) sector contributes to almost 2% of the global carbon footprint with an estimated trend of 3-3.6% by 2020. Most of this growth (45%) can be attributed to data centers (DC) which now represent the core infrastructure for different industries. Furthermore, cloud DCs are complex systems composed of several ICT and non-ICT (i.e. mechanical and electrical) sub-systems. The variety of configurations and the inter-dependencies of the different DC sub-systems leads to enormous challenges in understanding and optimizing DC energy efficiency based on the Power Usage Effectiveness (PUE) metric. Within this context, we focus in this work on analyzing the behavior of Deep Neural Network (DNN)-based model to predict the DC energy efficiency metric (PUE). In fact, the proposed model is used to evaluate the impact of various DC sub-systems on energy efficiency. Through an experimentation with real datasets from a real DC, we observed that DNN-based model achieves a good Root Mean Square Error (RMSE). The obtained results of this experimentation indicate that our proposed DNN-based model improves the PUE optimization, and consequently, shows its promise for a practical implementation.

In this paper, we present our proof-of-concept of a workload aware storage platform. The POC demonstrates the feasibility of building a machine learning technics facilitated middleware for storage management. The middleware is capable of providing optimal assignments of storage workloads to backends as well as continuously on-the-fly optimization thereafter. Experiment indicates that the proposed middleware can efficiently and dynamically adapt the storage backend to satisfy the SLA requirements with minimum impact on the workloads.

OpenStack does not provide high availability at application and VM (Virtual Machine) level. Some basic functionalities are provided by OpenStack Heat to restart/rebuild the VM/stack if it is not responsive. However, it does not support failover or different types of redundancy models. HAStack is a custom designed OpenStack Heat plugin in form of a multi-agent system which is intended to provide such functionalities by utilizing the PaceMaker (or OpenSAF) and OpenStack Nova API without touching the OpenStack code. These features will ensure HAStack full compatibility with OpenStack and ease of upgrade to a new version of OpenStack.

In this work, we present a novel approach that leverages Linux containers to achieve High Availability (HA) for cloud applications. A middleware that is comprised of a set of HA agents is defined to compensate the limitations of Linux containers in achieving HA. In our approach we start modeling at the application level, considering the dependencies among application components. We generate the proper scheduling scheme and then deploy the application across containers in the cloud. For each container that hosts critical component(s), we continuously monitor its status and checkpoint its full state, and then react to its failure by restarting locally or failing over to another host where we resume the computing from the most recent state. By using this strategy, all components hosted in a container are preserved without intrusively imposing modification on the application side. Finally, the feasibility of our approach is verified by building a proof-of-concept prototype and a case study of a video streaming application.

The design of avionic systems is a complex engineering activity. The iterative integration approach helps in controlling the complexity of such activity. On the other hand, using such approach to design evolving systems requires the reconfiguration of scheduling parameters of already integrated parts. This reconfiguration results in a recertification process having a cost that depends on the criticality level of the affected application. We propose a new approach which helps the system designer at each integration step in establishing the new scheduling parameters that minimize such cost. In this work, we focus on the Integrated Modular Avionic (IMA) architecture connected through a Time-Triggered Ethernet (TTEthernet) network. We present a formal model for such systems and we use this model to define a set of constraints that ensure the real-time requirements. These constraints are expressed using an SMTbased language and we used the SMT-solver YICES to find automatically a feasible scheduling parameters that minimize the cost of integration. We show our framework at work by analyzing the iterative integration of some functionalities of the Flight Management System.

Measuring the performance of cloud computing-based applications using ISO quality characteristics is a complex activity for various reasons, among them the complexity of the typical cloud computing infrastructure on which an application operates. To address this issue, the authors use Bautista's proposed performance measurement framework [1] on log data from an actual data centre to map and statistically analyze one of the ISO quality characteristics: time behavior. This empirical case study was conducted on an industry private cloud. The results of the study demonstrate that it is possible to use the proposed performance measurement framework in a cloud computing context. They also show that the framework holds great promise for expanding the experimentation to other ISO quality characteristics, larger volumes of data, and other statistical techniques that could be used to analyze performance.

Service availability is a key non-functional requirement that system architects and integrators seek to achieve. High Availability (HA) can be attained using a dedicated distributed HA middleware that can manage clustered redundant resources to maintain the continuous service delivery even in the presence of failures. Nonetheless, employing such middleware based solutions requires deep knowledge of the domain and substantial configuration effort, which can be a tedious and error prone task. In this paper, we demonstrate an automated approach that mitigates the efforts of configuring HA systems, and allows non-domain experts to easily use specialized HA solutions to increase the reliability of the services provided by their systems.

Cloud computing is a computing model which is essentially characterized by an on-demand and dynamic provisioning of computing resources. In this model, a cloud is a large-scale distributed system which leverages internet and virtualization technologies to provide computing resources as a service. Efficient, flexible and dynamic resource management is among the most challenging research issues in this domain. In this context, we present a study focusing on the dynamic behavior of the scheduling functionality of an Infrastructure-as-a-Service (IaaS) cloud, namely OpenStack Scheduler. We aim, through this study at identifying the limitations of this scheduler and ultimately enabling its extension using enhanced metrics. Towards this end, we present a Design of Experiment (DOE) based approach for the evaluation of the OpenStack Scheduler behavior. In particular, we use the screening type of experiment to identify the factors with significant effects on the responses. In our context, these factors are the amount of memory and the number of CPU cores assigned to virtual machine (VM) and the amount of memory and the number of cores on physical nodes. More specifically, we present a two-level fractional factorial balanced with the resolution IV and four center points experimental design with no replication.

The Availability Management Framework (AMF) is a service responsible for managing the availability of services provided by applications that run under its control. Standardized by the Service Availability Forum (SAF), AMF requires for its operations a complete and compliant AMF configuration of the applications to be managed. In this paper, we describe two complementary and integrated tools for AMF configurations generation and validation. Indeed, writing manually an AMF configuration is a tedious and error prone task as a large number of requirements defined in the standard have to be taken into consideration during the process. One solution for ensuring compliance with the standard is the validation of the configurations against all the AMF requirements. For this, we have designed and implemented a domain model for AMF configurations and use it as a basis for an AMF configuration validator. To further ease the task of a configuration designer, we have devised and implemented a method for generating automatically AMF configurations.

UML can be effectively used for the design and analysis of real-time systems. UML profiles for real-time, like the current OMG's standard UML/SPT, enable the modeling of quantitative requirements such as time constraints as well as behavioral features such as concurrency. Because of the multitude of diagrams and their complexity, UML/SPT models face the challenging issue of consistency. In this paper, we look into the behavioral consistency and particularly into concurrency-related properties of UML/SPT models. In order to do so, we formally define the UML/SPT concurrency domain model in terms of timed automata. As a straightforward application of this semantics, UML/SPT concurrent models can be validated using well-established model checking techniques and tools

UML through its profiling mechanism is well adapted for the modeling of real-time software requirements and designs. It is becoming the de facto standard. On the other hand, several real-time schedulability analysis techniques have been proposed in the literature. One of the current research challenges is in bridging the gap between the UML/SPT models and the well-established real-time schedulability analysis techniques. In this paper, we propose an MDA-compliant approach addressing this issue. We develop an UML metamodel for a well-established schedulability analysis technique. We propose a rule-based transformation between the UML/SPT schedulability analysis sub-profile metamodel and this metamodel. We illustrate our approach with an example