SURF Research Bootcamp presenteert een deep dive in onderzoeksvaardigheden in de vorm van vijf praktische workshops. De informatieve workshops zijn bedoeld voor onderzoekers en onderzoeksondersteuners. Bij alle sessies is een laptop vereist.
On June 27th, 2024, SURF along with Dutch Hardware Acceleration network organized the third edition of FIRE (FPGA Innovation Research Exchange) workshop. The theme of this edition was about tools and hardware acceleration for AI workloads. The event featured a lineup of discussions centered around cutting-edge computing technologies for AI workload acceleration, showcasing multiple industrial players such as Innatera and Groq as well as developers of HW acceleration platforms such as AMD and Altera. In addition, the workshop featured two keynote talks on future computing platforms dedicated to quantum computing as well as cellular automata.
The Julia programming language is being used more and more for high-performance computing and data processing (including on Snellius), due to being a combination of a high-level language with optimized native code generation. It also offers interesting interactive workflows that are not very common when working on high-performance code.
The JuliaCon conference is the annual multi-day conference on everything Julia. JuliaCon 2023 featured over 200 presentation sessions and many workshops. This year Julia conference is held in Eindhoven, 9-13 july 2024. This might be a nice opportunity for current users of Julia for HPC, or those interested in Julia, to get a feel for the Julia language and ecosystem, plus to get in touch with the Julia community. Early bird tickets are now available.
On December 6th, 2023, SURF in Utrecht successfully organized the second edition of FIRE FPGA symposium along with the hardware acceleration community. This is the continuation of the “FIRE: FPGA Innovation Research Exchange“ event series. The focus of this event was on application development using HW-SW co-design approaches. The event addressed the challenge of optimizing applications on heterogeneous CPU-accelerator systems to achieve full system optimization.
We had a number of speakers from industry, academia and research institutes. In addition, we organized a session with 5-minute flash presentations, where many researchers and companies presented their innovations and products at a high pace to engender collaboration and information exchange.
Ontdek trends en kom in contact met andere vakgenoten in HPC, AI, Machine Learning, Quantum en Data Science op 7 december in het Koninklijk Instituut voor de Tropen, Amsterdam. De toegang is gratis.
Hardware architecture improvements had relied on Moore's law and Dennard scaling for some decades. According to Moore's prediction, the number of transistors on a chip doubles almost every 18 months. According to Dennard scaling, as transistors become smaller, their power density stays constant. These allowed manufacturers to raise clock frequencies without significantly increasing overall circuit power consumption. Nowadays, Moore’s law is not happening as predicted, and Dennard scaling law is broken down. To address this, (homogeneous) multi-processors were invented, and opened a new era for parallel processing to gain better performance with a reasonable increase in power consumption. However, homogeneous multi-processors cannot keep up with high-demand application requirements such as low power, energy efficiency, and high performance.
In the ever-evolving realm of scientific research, the advent of Artificial Intelligence (AI) has been like the discovery of a new frontier. Yet, it isn't without its hurdles. This was a central topic during our panel session at SURF Research Day 2023, where we delved into the intricacies of employing machine learning as a research methodology.
Energy is an emerging topic in the scientific computing ecosystem and is becoming a design point for future research. Science relies increasingly on digital research computing as a tool for analysis and experimentation. The exponential increase in demand for computing means that classically designed ICT infrastructure will soon become unsustainable in terms of its energy footprint. We need to experiment with energy-efficient methods, tools, algorithms and hardware technologies. In the Netherlands, we are working towards zero energy waste for high-performance computing (HPC) applications on the national supercomputer “Snellius”. It involves discussing challenges, proposing new research directions, finding opportunities to engage the user community, and taking steps for the responsible use of software in research.
Traditionally, supercomputing focuses on improving latency or throughput, which are of massive importance for applications such as drug discovery or climate simulations. For many decades we developed infrastructure, algorithms, and software tools to obtain improvements. Given the rapid increase in energy usage for ICT services, further emphasised by the imminent energy crisis, it is a priority to understand and optimise the energy consumption of research computing applications
To know more, read the publication :
https://ercim-news.ercim.eu/en131/special/making-scientific-research-on…
Artificial Intelligence (AI) has revolutionized the research landscape, offering new methodologies and tools that have the potential to greatly enhance scientific discoveries. However, the adoption of AI in research also brings its own set of challenges, requiring new skills and a deep understanding of responsible AI practices. We need your insights to help us navigate this exciting yet complex field. Help us by filling in our 10-minute survey on responsible AI in research!
"Long-term society’s willingness to invest in HPC will depend on our ability to solve the most important societal and scientific problems, not in the ability to execute a subset of calculations that scale well to a billion of cores in a single run"
To solve challenges on a global scale, we need access to extraordinary computing capacity but we also need to build expertise, develop new approaches, explore new applications, etc. EuroHPC, based in Luxembourg, aims to achieve this in the near-term future. Its objective is to establish European leadership in HPC by jointly procuring exascale & pre-exascale supercomputers, managing open calls for application development and funding research and education collaboration across Europe.
