Heterogeneous computing systems provide high performance and energy efficiency. However, to optimally utilize such systems, solutions that distribute the work across host CPUs and accelerating devices are needed. In this paper, we present a performance and energy aware approach that combines AI planning heuristics for parameter space exploration with a machine learning model for performance and energy evaluation to determine a near-optimal system configuration. For data-parallel applications our approach determines a near-optimal host-device distribution of work, number of processing units required and the corresponding scheduling strategy. We evaluate our approach for various heterogeneous systems accelerated with GPU or the Intel Xeon Phi. The experimental results demonstrate that our approach finds a near-optimal system configuration by evaluating only about 7% of reasonable configurations. Furthermore, the performance per Joule estimation of system configurations using our machine learning model is more than 1000 x faster compared to the system evaluation by program execution.

A major challenge in modelling and simulation is the need to combine expertise in both software technologies and a given scientific domain. When High-Performance Computing (HPC) is required to solve a scientific problem, software development becomes a problematic issue. Considering the complexity of the software for HPC, it is useful to identify programming languages that can be used to alleviate this issue. Because the existing literature on the topic of HPC is very dispersed, we performed a Systematic Mapping Study (SMS) in the context of the European COST Action cHiPSet. This literature study maps characteristics of various programming languages for data-intensive HPC applications, including category, typical user profiles, effectiveness, and type of articles. We organised the SMS in two phases. In the first phase, relevant articles are identified employing an automated keyword-based search in eight digital libraries. This lead to an initial sample of 420 papers, which was then narrowed down in a second phase by human inspection of article abstracts, titles and keywords to 152 relevant articles published in the period 2006–2018. The analysis of these articles enabled us to identify 26 programming languages referred to in 33 of relevant articles. We compared the outcome of the mapping study with results of our questionnaire-based survey that involved 57 HPC experts. The mapping study and the survey revealed that the desired features of programming languages for data-intensive HPC applications are portability, performance and usability. Furthermore, we observed that the majority of the programming languages used in the context of data-intensive HPC applications are text-based general-purpose programming languages. Typically these have a steep learning curve, which makes them difficult to adopt. We believe that the outcome of this study will inspire future research and development in programming languages for data-intensive HPC applications.

Deep learning is an important component of big-data analytic tools and intelligent applications, such as, self-driving cars, computer vision, speech recognition, or precision medicine. However, the training process is computationally intensive, and often requires a large amount of time if performed sequentially. Modern parallel computing systems provide the capability to reduce the required training time of deep neural networks.In this paper, we present our parallelization scheme for training convolutional neural networks (CNN) named Controlled Hogwild with Arbitrary Order of Synchronization (CHAOS). Major features of CHAOS include the support for thread and vector parallelism, non-instant updates of weight parameters during back-propagation without a significant delay, and implicit synchronization in arbitrary order. CHAOS is tailored for parallel computing systems that are accelerated with the Intel Xeon Phi. We evaluate our parallelization approach empirically using measurement techniques and performance modeling for various numbers of threads and CNN architectures. Experimental results for the MNIST dataset of handwritten digits using the total number of threads on the Xeon Phi show speedups of up to 103x compared to the execution on one thread of the Xeon Phi, 14x compared to the sequential execution on Intel Xeon E5, and 58x compared to the sequential execution on Intel Core i5.

Determining the optimal location of control cabinet components requires the exploration of a large configuration space. For real-world control cabinets it is impractical to evaluate all possible cabinet configurations. Therefore, we need to apply methods for intelligent exploration of cabinet configuration space that enable to find a near-optimal configuration without evaluation of all possible configurations.In this paper, we describe an approach for multi-objective optimization of control cabinet layout that is based on Pareto Simulated Annealing. Optimization aims at minimizing the total wire length used for interconnection of components and the heat convection within the cabinet. We simulate heat convection to study the warm air flow within the control cabinet and determine the optimal position of components that generate heat during the operation. We evaluate and demonstrate the effectiveness of our approach empirically for various control cabinet sizes and usage scenarios.

Noise is any undesired environmental sound. A sound at the same dB level may be perceived as annoying noise or as pleasant music. Therefore, it is necessary to go beyond the state-of-the-art approaches that measure only the dB level and also identify the type of noise. In this paper, we present a machine learning based method for urban noise identification using an inexpensive IoT unit. We use Mel-frequency cepstral coefficients for audio feature extraction and supervised classification algorithms (that is, support vector machine, k-nearest neighbors, bootstrap aggregation, and random forest) for noise classification. We evaluate our approach experimentally with a data-set of about 3000 sound samples grouped in eight sound classes (such as car horn, jackhammer, or street music). We explore the parameter space of the four algorithms to estimate the optimal parameter values for classification of sound samples in the data-set under study. We achieve a noise classification accuracy in the range 88% - 94%.

We present IoTutor that is a cognitive computing solution for education of students in the IoT domain. We implement the IoTutor as a platform-independent web-based application that is able to interact with users via text or speech using natural language. We train the IoTutor with selected scientific publications relevant to the IoT education. To investigate users' experience with the IoTutor, we ask a group of students taking an IoT master level course at the Linnaeus University to use the IoTutor for a period of two weeks. We ask students to express their opinions with respect to the attractiveness, perspicuity, efficiency, stimulation, and novelty of the IoTutor. The evaluation results show a trend that students express an overall positive attitude towards the IoTutor with majority of the aspects rated higher than the neutral value.

Recent developments in sensor technology, wearable computing, Internet of Things (IoT), and wireless communication have given rise to research in ubiquitous healthcare and remote monitoring of human’s health and activities. Health monitoring systems involve processing and analysis of data retrieved from smartphones, smart watches, smart bracelets, as well as various sensors and wearable devices. Such systems enable continuous monitoring of patients psychological and health conditions by sensing and transmitting measurements such as heart rate, electrocardiogram, body temperature, respiratory rate, chest sounds, or blood pressure. Pervasive healthcare, as a relevant application domain in this context, aims at revolutionizing the delivery of medical services through a medical assistive environment and facilitates the independent living of patients. In this chapter, we discuss (1) data collection, fusion, ownership and privacy issues; (2) models, technologies and solutions for medical data processing and analysis; (3) big medical data analytics for remote health monitoring; (4) research challenges and opportunities in medical data analytics; (5) examples of case studies and practical solutions.

Many complex problems, such as natural language processing or visual object detection, are solved using deep learning. However, efficient training of complex deep convolutional neural networks for large data sets is computationally demanding and requires parallel computing resources. In this paper, we present two parameterized performance models for estimation of execution time of training convolutional neural networks on the Intel many integrated core architecture. While for the first performance model we minimally use measurement techniques for parameter value estimation, in the second model we estimate more parameters based on measurements. We evaluate the prediction accuracy of performance models in the context of training three different convolutional neural network architectures on the Intel Xeon Phi. The achieved average performance prediction accuracy is about 15% for the first model and 11% for second model.

Scientific and technological innovations have become increasingly important as we face the benefits and challenges of both globalization and a knowledge-based economy. Still, enrolment rates in STEM degrees are low in many European countries and consequently there is a lack of adequately educated workforce in industries. We believe that this can be mainly attributed to pedagogical issues, such as the lack of engaging hands-on activities utilized for science and math education in middle and high schools. In this paper, we report our work in the SciChallenge European project, which aims at increasing the interest of pre-university students in STEM disciplines, through its distinguishing feature, the systematic use of social media for providing and evaluation of the student-generated content. A social media-aware contest and platform were thus developed and tested in a pan-European contest that attracted >700 participants. The statistical analysis and results revealed that the platform and contest positively influenced participants STEM learning and motivation, while only the gender factor for the younger study group appeared to affect the outcomes (confidence level – p<.05).