Energy-aware High Performance Multi-objective Optimization in Heterogeneous Computer Architectures. Applications on Biomedical (e-hpMOBE)
Storage technology and distributed platforms, including networks, sensors and other data collection devices have enabled the generation of large databases, from which, enormous socio-economic interest applications are emerging. The integration of systems with increasing processing and communications capabilities in a great variety of everyday devices allows the development of applications with very different profiles in terms of requirements on speed, consumption, portability, etc., and have also given rise to paradigms as the Internet of Things (IoT), Cloud Computing, or Big Data, which involve significant changes in terms of how we interact with our environment and also how we access to information and communications technologies.
This project assumes that applications for analyzing large volumes of high dimensional data are not possible without the efficient use, in terms of performance and power consumption, of parallel architectures and distributed heterogeneous computers, including accelerators such as GPUs, and storage resources managed by distributed file systems. Thus, taking the results of the TIN2012-32039 project as the starting basis, relevant contributions will be provided parallel and energy efficient multi-objective optimization and applications on neuroengineering, rehabilitation technologies and medical images processing.