TASK Quarterly

Energy-Efficiency Support Services on the CAISE Platform

Fri, 14 Aug 2026 00:00:00 +0200

The article presents a package of services supporting energy efficiency within the CAISE platform environment, aimed at monitoring, managing, and optimizing energy consumption in cloud computing environments. The monitoring services include Computation Energy Consumption Monitoring (MEO) and Computation Intensity Monitoring (MIO), which enable real-time tracking of energy usage parameters and resource load. The management and administrative layer is formed by System Energy Configuration (KES) and Optimization Parameter Logging (RPO), which allow the definition of energy policies and the collection of data relevant for subsequent analysis and system tuning. A key component of the package is the Performance–Energy Optimization (OWE) service, which supports the selection of configurations and execution methods for computational tasks in order to achieve a balance between performance and energy efficiency. The paper discusses the functional assumptions, supported hardware–software configurations, and implementation aspects of these services. It also outlines the requirements and challenges related to their practical deployment, potential application perspectives and use cases in large-scale computing environments, as well as selected practical results for OWE.

Functionality comparison of open-source PaaS monitoring systems

Filip Sołtys, Henryk Krawczyk — Fri, 14 Aug 2026 00:00:00 +0200

For Platform-as-a-Service (PaaS) environments characterized by elasticity, multi-tenancy, and layered dependencies, this paper proposes a structured method for selecting a vendor-neutral open-source observability stack. The proposed method is based on an analysis of commercial monitoring solutions and covers five functional categories: metrics monitoring, visualization, log monitoring, distributed tracing, and alert routing/incident response. Selected open-source tools are evaluated using technical and operational criteria, including integration, scalability, high availability, and maintainability, with a 1–5 Mean Opinion Score (MOS)-style rubric. Based on the results, the proposed toolchain consists of Prometheus, Grafana, Loki, Jaeger, and Alertmanager. It can serve as a practical monitoring framework for different PaaS environments.

Systematic Evaluation of 2D and 3D Deep Learning Detectors for Pulmonary Nodule Detection in Chest CT

Fri, 14 Aug 2026 00:00:00 +0200

Accurate detection of pulmonary nodules in chest CT is essential for early lung cancer diagnosis. This study presents a systematic evaluation of two-dimensional (2D) and three-dimensional (3D) deep learning detectors for pulmonary nodule detection using the LUNA16 benchmark dataset. While 3D deep learning approaches can exploit full volumetric information, 2D detectors remain attractive due to their lower computational complexity and simpler training requirements. Particular attention was devoted to the effect of annotation preprocessing in the 2D setting. Two 2D detectors (RetinaNet and YOLO) and two 3D detectors (3D RetinaNet and a hybrid CNN-Transformer architecture) were evaluated. For the 2D experiments, multiple dataset variants were prepared by filtering annotations according to minimum bounding-box area and by balancing positive and negative slices. Detection performance was
evaluated using F1-score, precision, recall, mAP, mAR, and FROC-based metrics across multiple IoU thresholds. The results showed that preprocessing strategies substantially affected 2D detector performance. Filtering very small peripheral annotations and balancing the datasets improved localisation quality and detector stability for both RetinaNet and YOLO. Among the 2D configurations, the best results were obtained for the balanced filtered subsets. The 3D RetinaNet achieved the strongest overall performance, providing higher detection sensitivity, more stable localisation, and better generalisation between validation and independent test data. The CNN-Transformer also benefited from volumetric input and retained relatively high recall at lower IoU thresholds, although its performance
remained lower than that of the 3D RetinaNet. The findings demonstrate the importance of volumetric context for robust pulmonary nodule detection and indicate that appropriate annotation preprocessing can substantially improve the effectiveness of slice-based 2D detection frameworks.