<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>ESA |</title><link>https://example.com/tags/esa/</link><atom:link href="https://example.com/tags/esa/index.xml" rel="self" type="application/rss+xml"/><description>ESA</description><generator>HugoBlox Kit (https://hugoblox.com)</generator><language>en-us</language><lastBuildDate>Mon, 01 Jan 2018 00:00:00 +0000</lastBuildDate><image><url>https://example.com/media/icon_hu_eee4a95885829ab2.png</url><title>ESA</title><link>https://example.com/tags/esa/</link></image><item><title>GPU4S</title><link>https://example.com/projects/gpu4s/</link><pubDate>Mon, 01 Jan 2018 00:00:00 +0000</pubDate><guid>https://example.com/projects/gpu4s/</guid><description>&lt;p&gt;&lt;strong&gt;GPU4S&lt;/strong&gt; (&lt;em&gt;Embedded GPUs for Space&lt;/em&gt;) is a research initiative funded by the &lt;strong&gt;European Space Agency (ESA)&lt;/strong&gt; evaluating commercial off-the-shelf (COTS) embedded graphics processing units for space payload processing and autonomous navigation.&lt;/p&gt;
&lt;h3 id="-objectives--scope"&gt;🎯 Objectives &amp;amp; Scope&lt;/h3&gt;
&lt;p&gt;Space applications increasingly require high-throughput parallel computing for satellite image processing, SAR, and autonomous navigation. GPU4S investigated the feasibility and reliability of deploying embedded GPUs in harsh radiation space environments.&lt;/p&gt;
&lt;h3 id="-key-research-innovations"&gt;💡 Key Research Innovations&lt;/h3&gt;
&lt;ul&gt;
&lt;li&gt;&lt;strong&gt;Space GPU Benchmark Suite:&lt;/strong&gt; Developed a standardized open-source benchmark suite representative of space payload processing workloads.&lt;/li&gt;
&lt;li&gt;&lt;strong&gt;SafeSoftDR (Software-Based Diverse Redundancy):&lt;/strong&gt; Created a software-only fault tolerance library providing kernel-level diverse redundancy on GPUs to mitigate soft errors without hardware modification.&lt;/li&gt;
&lt;li&gt;&lt;strong&gt;COTS Architecture Evaluation:&lt;/strong&gt; Evaluated low-power embedded GPU architectures (ARM Mali, NVIDIA Jetson, AMD) for radiation tolerance and performance-per-watt.&lt;/li&gt;
&lt;/ul&gt;
&lt;h3 id="-key-collaborators"&gt;🏛️ Key Collaborators&lt;/h3&gt;
&lt;ul&gt;
&lt;li&gt;&lt;strong&gt;Barcelona Supercomputing Center (BSC)&lt;/strong&gt;&lt;/li&gt;
&lt;li&gt;&lt;strong&gt;European Space Agency (ESA)&lt;/strong&gt;&lt;/li&gt;
&lt;li&gt;&lt;strong&gt;Airbus Defence and Space&lt;/strong&gt;&lt;/li&gt;
&lt;/ul&gt;</description></item></channel></rss>