Energy of computing is a serious environmental concern and mitigating it is an important technological challenge. Accurate measurement of energy consumption during an application execution is key to application-level energy minimization techniques. There are three popular approaches to providing it: (a) System-level physical measurements using external power meters; (b) Measurements using on-chip power sensors and (c) Energy predictive models. In this work, we present a comprehensive study comparing the accuracy of state-of-the-art on-chip power sensors and energy predictive models against system-level physical measurements using external power meters, which we consider to be the ground truth. We show that the average error of the dynamic energy profiles obtained using on-chip power sensors can be as high as 73% and the maximum reaches 300% for two scientific applications, matrix-matrix multiplication and 2D fast Fourier transform for a wide range of problem sizes. The applications are executed on three modern Intel multicore CPUs, two Nvidia GPUs and an Intel Xeon Phi accelerator. The average error of the energy predictive models employing performance monitoring counters (PMCs) as predictor variables can be as high as 32% and the maximum reaches 100% for a diverse set of seventeen benchmarks executed on two Intel multicore CPUs (one Haswell and the other Skylake). We also demonstrate that using inaccurate energy measurements provided by on-chip sensors for dynamic energy optimization can result in significant energy losses up to 84%. We show that, owing to the nature of the deviations of the energy measurements provided by on-chip sensors from the ground truth, calibration can not improve the accuracy of the on-chip sensors to an extent that can allow them to be used in optimization of applications for dynamic energy. Finally, we present the lessons learned, our recommendations for the use of on-chip sensors and energy predictive models and future directions.

Performance and energy are the two most important objectives for optimization on modern parallel platforms. In this article, we show that moving from single-objective optimization for performance or energy to their bi-objective optimization on heterogeneous processors results in a tremendous increase in the number of optimal solutions (workload distributions) even for the simple case of linear performance and energy profiles. We then study full performance and energy profiles of two real-life data-parallel applications and find that they exhibit shapes that are non-linear and complex enough to prevent good approximation of them as analytical functions for input to exact algorithms or optimization software for determining the Pareto front. We, therefore, propose a solution method solving the bi-objective optimization problem on heterogeneous processors. The method's novel component is an efficient and exact global optimization algorithm that takes as an input performance and energy profiles as arbitrary discrete functions of workload size, which accurately and realistically take into account resource contention and NUMA inherent in modern parallel platforms, and returns the Pareto-optimal solutions (generally speaking, load imbalanced). To construct the input discrete energy functions, the method employs a methodology that accurately models the energy consumption by a hybrid data-parallel application executing on a heterogeneous HPC platform containing different computing devices using system-level power measurements provided by power meters. We experimentally analyse the proposed solution method using three data-parallel applications, matrix multiplication, 2D fast Fourier transform (2D-FFT), and gene sequencing, on two connected heterogeneous servers consisting of multicore CPUs, GPUs, and Intel Xeon Phi. We show that it determines a superior Pareto front containing the best load balanced solutions and all the load imbalanced solutions that are ignored by load balancing methods.

Identification and re-identification are two major security and privacy threats to medical imaging data. De-identification in DICOM medical data is essential to preserve the privacy of patients’ Personally Identifiable Information (PII) and requires a systematic approach. However, there is a lack of sufficient detail regarding the de-identification process of DICOM attributes, for example, what needs to be considered before removing a DICOM attribute. In this paper, we first highlight and review the key challenges in the medical image data de-identification process. In this paper, we develop a two-stage de-identification process for CT scan images available in DICOM file format. In the first stage of the de-identification process, the patient’s PII—including name, date of birth, etc., are removed at the hospital facility using the export process available in their Picture Archiving and Communication System (PACS). The second stage employs the proposed DICOM de-identification tool for an exhaustive attribute-level investigation to further de-identify and ensure that all PII has been removed. Finally, we provide a roadmap for future considerations to build a semi-automated or automated tool for the DICOM datasets de-identification.

Obesity is a major public health problem worldwide, and the prevalence of childhood obesity is of particular concern. Effective interventions for preventing and treating childhood obesity aim to change behaviour and exposure at the individual, community, and societal levels. However, monitoring and evaluating such changes is very challenging. The EU Horizon 2020 project “Big Data against Childhood Obesity (BigO)” aims at gathering large-scale data from a large number of children using different sensor technologies to create comprehensive obesity prevalence models for data-driven predictions about specific policies on a community. It further provides real-time monitoring of the population responses, supported by meaningful real-time data analysis and visualisations. Since BigO involves monitoring and storing of personal data related to the behaviours of a potentially vulnerable population, the data representation, security, and access control are crucial. In this paper, we briefly present the BigO system architecture and focus on the necessary components of the system that deals with data access control, storage, anonymisation, and the corresponding interfaces with the rest of the system. We propose a three-layered data warehouse architecture: The back-end layer consists of a database management system for data collection, de-identification, and anonymisation of the original datasets. The role-based permissions and secured views are implemented in the access control layer. Lastly, the controller layer regulates the data access protocols for any data access and data analysis. We further present the data representation methods and the storage models considering the privacy and security mechanisms. The data privacy and security plans are devised based on the types of collected personal, the types of users, data storage, data transmission, and data analysis. We discuss in detail the challenges of privacy protection in this large distributed data-driven application and implement novel privacy-aware data analysis protocols to ensure that the proposed models guarantee the privacy and security of datasets. Finally, we present the BigO system architecture and its implementation that integrates privacy-aware protocols.

Open-source automated insulin delivery (AID) technologies use the latest continuous glucose monitors (CGM), insulin pumps, and algorithms to automate insulin delivery for effective diabetes management. Early community-wide adoption of open-source AID, such as OpenAPS, has motivated clinical and research communities to understand and evaluate glucose-related outcomes of such user-driven innovation. Initial OpenAPS studies include retrospective studies assessing high-level outcomes of average glucose levels and HbA1c, without in-depth analysis of glucose variability (GV). The OpenAPS Data Commons dataset, donated to by open-source AID users with insulin-requiring diabetes, is the largest freely available diabetes-related dataset with over 46,070 days’ worth of data and over 10 million CGM data points, alongside insulin dosing and algorithmic decision data. This paper first reviews the development toward the latest open-source AID and the performance of clinically approved GV metrics. We evaluate the GV outcomes using large-scale data analytics for the n = 122 version of the OpenAPS Data Commons. We describe the data cleaning processes, methods for measuring GV, and the results of data analysis based on individual self-reported demographics. Furthermore, we highlight the lessons learned from the GV outcomes and the analysis of a rich and complex diabetes dataset and additional research questions that emerged from this work to guide future research. This paper affirms previous studies’ findings of the efficacy of open-source AID.

Conversational AI systems have emerged as key enablers of human-like interactions across diverse sectors. Nevertheless, the balance between linguistic nuance and factual accuracy has proven elusive. In this paper, we first introduce LLMXplorer, a comprehensive tool that provides an in-depth review of over 205 large language models (LLMs), elucidating their practical implications, ranging from social and ethical to regulatory, as well as their applicability across industries. Building on this foundation, we propose a novel functional architecture that seamlessly integrates the structured dynamics of knowledge graphs with the linguistic capabilities of LLMs. Validated using real-world AI news data, our architecture adeptly blends linguistic sophistication with factual rigor and further strengthens data security through role-based access control. This research provides insights into the evolving landscape of conversational AI, emphasizing the imperative for systems that are efficient, transparent, and trustworthy.