Multi-GPU nodes are increasingly common in the rapidly evolving landscape of exascale supercomputers. On these systems, GPUs on the same node are connected through dedicated networks, with bandwidths up to a few terabits per second. However, gauging performance expectations and maximizing system efficiency is challenging due to different technologies, design options, and software layers. This paper comprehensively characterizes three supercomputers — Alps, Leonardo, and LUMI — each with a unique architecture and design. We focus on performance evaluation of intra-node and inter-node interconnects on up to 4,096 GPUs, using a mix of intra-node and inter-node benchmarks. By analyzing its limitations and opportunities, we aim to offer practical guidance to researchers, system architects, and software developers dealing with multi-GPU supercomputing. Our results show that there is untapped bandwidth, and there are still many opportunities for optimization, ranging from network to software optimization.

Image registration is a well-defined computation paradigm widely applied to align one or more images to a target image. This paradigm, which builds upon three main components, is particularly compute-intensive and represents many image processing pipelines’ bottlenecks. State-of-the-art solutions leverage hardware acceleration to speed up image registration, but they are usually limited to implementing a single component. We present Faber, an open-source HW/SW CAD toolchain tailored to image registration. The Faber toolchain comprises HW/SW highly-tunable registration components, supports users with different expertise in building custom pipelines, and automates the design process. In this direction, Faber provides both default settings for entry-level users and latency and resource models to guide HW experts in customizing the different components. Finally, Faber achieves from 1.5× to 54× in speedup and from 2× to 177× in energy efficiency against state-of-the-art tools on a Xeon Gold.

We present the results of the first experiment to ¿plug¿ the Small BAseline Subset (SBAS) DInSAR algorithm into a GRID-based system; the key idea is to combine the robustness of the exploited advanced interferometric SAR approach with the high computing capability provided by a GRID environment. In particular, we have exploited the low-resolution SBAS algorithm and we benefited of the availability of the ESA Grid Processing-on-demand environment. The presented results, carried out on ENVISAT data, provide a overview of the main characteristics of the implemented SBAS-GRID processing solution.

We propose the use of MSG images for local ground radiance estimation. To this end, first the presence of clouds must be detected, then the satellite measurements must be correlated with measurements at ground. A Cloud Detection Net (CDN) algorithm has been implemented for the first task, while time series of radiance measured at ground by the National Research Council's Institute of Atmospheric Sciences and Climate have been compared with the MSG radiance measures. The joint use of ground and satellite radiance values led to the development of a Radiance Estimation Net (REN) algorithm yielding space-based estimates of the radiance on the field.

In continuation of the BACCHUS project, aimed at establishing a reference high quality geographic information system for vineyards, an airborne SAR survey has been carried out in fall 2005 in the Frascati area, near Rome (Italy) to demonstrate the potential of airborne radar remote sensing in vineyard characterization. This contribution reports on the polarimetric L-band and dual polarization C-band SAR data acquisition campaign supported by ESA and carried out on two dates in October 2005 (the first during the grape harvest and the other after the vintage completion).

This paper provides an overview of the project of the European Space Agency Building Environment for carried out in collaboration with UNESCO in support of the World heritage Convention. The project aims at demonstrate the capabilities of Earth Observation (EO) technology to provide reliable, synoptic and continuous information to Natural Park managers, Non-Governmental Organizations (NGOs) and local authorities to monitor and manage world heritage natural sites. This pilot project focuses on five different Parks in Central Africa, which correspond with the Gorilla habitat.