EuroHPC Highlights RED-SEA Project’s Role in Paving the Way for European Exascale Systems

July 8, 2024

July 8, 2024 — The EuroHPC JU-funded project, RED-SEA, has made substantial advancements in the development of scalable and efficient interconnect technologies that are essential for exascale computing.

Launched in April 2021, the RED-SEA project has, after three intense years, recently been concluded. The RED-SEA project, along with its sister projects DEEP-SEA and IO-SEA, has developed key technologies based on the Modular Supercomputing Architecture (MSA), which will help prepare Europe for the exascale era in supercomputing. The three projects worked together on a common, blueprint architecture that will support the deployment of highly efficient and scalable exascale systems.

The RED-SEA project brought together European academics and key European industrial enterprises in the domain of interconnection networks, with a total of 12 partners from six different countries: France, Germany, Greece, Italy, Spain, Switzerland. It had a budget of €8 000 000 from Horizon Europe, the EU’s funding programme for research and innovation.

EuroHPC JU interviewed Ms. Claire Chen, project coordinator of RED-SEA. Below, she highlights the key features of the project and explained what it has accomplished in the past three years.


EuroHPC: Can you please describe the RED-SEA project in your own words?

Claire Chen: Exascale supercomputers represent a significant leap in computing power and are made up processors which contain hundreds of thousands of nodes and millions of cores. With this technology, supercomputers are capable of performing a quintillion (1018) calculations per second. Every part of the supercomputer needs to work seamlessly together and to do this, they rely on an software that ensures that there is efficient distribution of tasks’ shared among all these different components.

Supercomputers use different types of processors such as GPUs for graphics and vector accelerators for extremely efficient supercomputing . These processors are essential for handling large amounts of data and running AI-related applications.

Interconnection networks are very important for the supercomputer’s performance. They serve as the backbone and handle many connections, support many parallel tasks, and work well with new data-focused and AI applications. They also need to be smart, managing resources such as energy efficiently and performing calculations within the network itself.

The RED-SEA project has focused on building the next generation of these interconnection network. It aimed to create a European network that is fast, can grow easily, and is reliable.

What have been the key objectives for the RED-SEA project and what progress has been made

Our project’s key objectives were supported by three pillars.

The first pillar focused on preparing the next-generation high-performance network interconnect technologies through specification and design. To this end, we leveraged existing European interconnect technologies like BullSequana eXascale Interconnect (BXI) and relevant Intellectual Properties (IPs) from previous EU-funded projects (e.g., Exanest).

The second pillar focused on exploring innovative network solutions.

The third pillar of our project aimed at developing an ecosystem of interconnect technologies with a community of users and developers. By bringing together research and industrial teams, we fostered collaboration and created a supportive environment for the advancement of interconnect technologies.

After three years of dedicated efforts, we have successfully advanced the state of interconnect network technologies and achieved these objectives.

One significant outcome lies in the advancement of the European Interconnect BXI, with a focus on enhancing the current version (BXIv2) and laying the groundwork for the next generation (BXIv3). Another key achievement has been the exploration of new, efficient network resource management. For example, we have enhanced network features such as collective operations by offloading collaborative work from compute resources (CPU, GPUs) to the network components. We have also improved congestion control to prevent overuse of network resources thus directly reducing global latencies.

Can you give some concrete examples of how RED-SEA project supports European HPC users and how it promotes greener and more sustainable supercomputing?

Sure! The RED-SEA project supported European HPC users and promoted greener and more sustainable supercomputing in several ways. Given the enormous power requirements of exascale computing, energy efficiency is a critical consideration for the interconnection network.

Our project has designed low-power communication protocols between the different supercomputer components and optimised network topology have contributed to minimise the overall energy consumption of the supercomputers.

We have also managed to find new techniques to offload collaborative work which will allow compute resources to focus on their primary tasks without actively managing network operations and to enable the network components to handle these operations independently. This has resulted in significant time and energy savings.

In addition to this, reducing congestion within the network enhances the efficiency of communications. In summary, we also focused on optimising the global performance of HPC systems, allowing users to accomplish more computational tasks with less energy consumption.

Our project has provided users with support in efficient and sustainable interconnect networks solutions, fostering collaboration, and driving innovation in green computing practices. Another aspect of this support is the use of simulation which is a popular method for evaluating the behavior and performance of IT systems such as HPC clusters. Simulation has been extensively used to model and assess new designs for high-performance interconnection networks. These simulation tools have been instrumental in designing next-generation interconnects within the project.

We also offered open-source tools that enable the collection of network traces and network simulators to reproduce the communication traffic of parallel applications in high-performance interconnection networks. By providing these tools, users, and developers can evaluate the performance of interconnect networks in their IT environments, thus enhancing their ability to innovate and optimise interconnect networks effectively.

This comprehensive support will allow users to leverage cutting-edge tools to drive advancements in HPC and interconnect network design.

What were the main challenges you encountered during the project’s development, if any?

The main challenges we faced was the development of a prototype of the new gateway. We explored and evaluated various technologies which generated a lot of changes in the design and increased the complexity of testing activities.

We also had to deal with changes in the project team, resource allocation and dissemination difficulties. However, we overcame these challenges thanks to proactive problem-solving, teamwork, and adaptability of everyone involved.

How is the development of the RED-SEA project supporting the ambition of the EuroHPC JU to make Europe a world-leader in supercomputing?

It is important to remember that interconnect networks play a pivotal role in the overall performance of Exascale systems and will serve as a backbone to future Exascale systems. By advancing high-performance interconnect technologies such as BXI, RED-SEA enhanced Europe’s capabilities in building exascale systems.

I believe that the RED-SEA collaboration is unique and strengthened Europe’s innovation ecosystem and its competitiveness in the global supercomputing landscape. Overall, I am proud to say that the progress made within the RED-SEA project aligns closely with EuroHPC JU’s objectives of establishing Europe as a world leader in supercomputing.

What’s next for your project and the results developed under this project?

We have identified and developed 21 RED-SEA exploitable results and two patents. Over half of these results are product-related, with some already integrated into commercial products. The other results should be integrated into commercial products within two years.

The outcomes of the RED-SEA project have also laid the groundwork for the specification and design of BXIv3, the next generation of high-performance network interface card designed for advanced computing environments, so hopefully it will be developed in future projects.


Next Steps

Following the conclusion of the RED-SEA project in March 2024, the EuroHPC Joint Undertaking is currently evaluating proposals submitted in response to the call for Innovation Action in Low Latency and High Bandwidth Interconnects.

This call will continue to advance European capabilities in HPC by developing a comprehensive roadmap for scalable inter-node interconnects, specifically targeting exascale and post-exascale systems.

The new proposals will build on the achievements of the RED-SEA project, incorporating the project’s findings and advancements. The call aimed at addressing the complete lifecycle of interconnect hardware and software, from design and development to testing and integration. The focus will also need to include features/capabilities ensuring high speed, low-delay, energy efficiency, virtualisation, scalability, reliability, and security.


Source: EuroHPC JU

Subscribe to HPCwire's Weekly Update!

Be the most informed person in the room! Stay ahead of the tech trends with industry updates delivered to you every week!

Aurora AI-Driven Atmosphere Model is 5,000x Faster Than Traditional Systems

July 16, 2024

While the onset of human-driven climate change brings with it many horrors, the increase in the frequency and strength of storms poses an enormous threat to communities across the globe. As climate change is warming ocea Read more…

Researchers Say Memory Bandwidth and NVLink Speeds in Hopper Not So Simple

July 15, 2024

Researchers measured the real-world bandwidth of Nvidia's Grace Hopper superchip, with the chip-to-chip interconnect results falling well short of theoretical claims. A paper published on July 10 by researchers in the U. Read more…

Belt-Tightening in Store for Most Federal FY25 Science Budets

July 15, 2024

If it’s summer, it’s federal budgeting time, not to mention an election year as well. There’s an excellent summary of the curent state of FY25 efforts reported in AIP’s policy FYI: Science Policy News. Belt-tight Read more…

Peter Shor Wins IEEE 2025 Shannon Award

July 15, 2024

Peter Shor, the MIT mathematician whose ‘Shor’s algorithm’ sent shivers of fear through the encryption community and helped galvanize ongoing efforts to build quantum computers, has been named the 2025 winner of th Read more…

Weekly Wire Roundup: July 8-July 12, 2024

July 12, 2024

HPC news can get pretty sleepy in June and July, but this week saw a bump in activity midweek as Americans realized they still had work to do after the previous holiday weekend. The world outside the United States also s Read more…

Nvidia, Intel not Welcomed in New Apple AI and HPC Development Tools

July 12, 2024

New Mac developer tools will leverage Apple's homegrown chips, limiting HPC users' ability to use parallel programming frameworks from Intel or Nvidia. Apple's latest programming framework, Xcode 16, was introduced at Read more…

Aurora AI-Driven Atmosphere Model is 5,000x Faster Than Traditional Systems

July 16, 2024

While the onset of human-driven climate change brings with it many horrors, the increase in the frequency and strength of storms poses an enormous threat to com Read more…

Shutterstock 1886124835

Researchers Say Memory Bandwidth and NVLink Speeds in Hopper Not So Simple

July 15, 2024

Researchers measured the real-world bandwidth of Nvidia's Grace Hopper superchip, with the chip-to-chip interconnect results falling well short of theoretical c Read more…

Shutterstock 2203611339

NSF Issues Next Solicitation and More Detail on National Quantum Virtual Laboratory

July 10, 2024

After percolating for roughly a year, NSF has issued the next solicitation for the National Quantum Virtual Lab program — this one focused on design and imple Read more…

NCSA’s SEAS Team Keeps APACE of AlphaFold2

July 9, 2024

High-performance computing (HPC) can often be challenging for researchers to use because it requires expertise in working with large datasets, scaling the softw Read more…

Anders Jensen on Europe’s Plan for AI-optimized Supercomputers, Welcoming the UK, and More

July 8, 2024

The recent ISC24 conference in Hamburg showcased LUMI and other leadership-class supercomputers co-funded by the EuroHPC Joint Undertaking (JU), including three Read more…

Generative AI to Account for 1.5% of World’s Power Consumption by 2029

July 8, 2024

Generative AI will take on a larger chunk of the world's power consumption to keep up with the hefty hardware requirements to run applications. "AI chips repres Read more…

US Senators Propose $32 Billion in Annual AI Spending, but Critics Remain Unconvinced

July 5, 2024

Senate leader, Chuck Schumer, and three colleagues want the US government to spend at least $32 billion annually by 2026 for non-defense related AI systems.  T Read more…

Point and Click HPC: High-Performance Desktops

July 3, 2024

Recently, an interesting paper appeared on Arvix called Use Cases for High-Performance Research Desktops. To be clear, the term desktop in this context does not Read more…

Atos Outlines Plans to Get Acquired, and a Path Forward

May 21, 2024

Atos – via its subsidiary Eviden – is the second major supercomputer maker outside of HPE, while others have largely dropped out. The lack of integrators and Atos' financial turmoil have the HPC market worried. If Atos goes under, HPE will be the only major option for building large-scale systems. Read more…

Everyone Except Nvidia Forms Ultra Accelerator Link (UALink) Consortium

May 30, 2024

Consider the GPU. An island of SIMD greatness that makes light work of matrix math. Originally designed to rapidly paint dots on a computer monitor, it was then Read more…

Comparing NVIDIA A100 and NVIDIA L40S: Which GPU is Ideal for AI and Graphics-Intensive Workloads?

October 30, 2023

With long lead times for the NVIDIA H100 and A100 GPUs, many organizations are looking at the new NVIDIA L40S GPU, which it’s a new GPU optimized for AI and g Read more…

Shutterstock_1687123447

Nvidia Economics: Make $5-$7 for Every $1 Spent on GPUs

June 30, 2024

Nvidia is saying that companies could make $5 to $7 for every $1 invested in GPUs over a four-year period. Customers are investing billions in new Nvidia hardwa Read more…

Nvidia Shipped 3.76 Million Data-center GPUs in 2023, According to Study

June 10, 2024

Nvidia had an explosive 2023 in data-center GPU shipments, which totaled roughly 3.76 million units, according to a study conducted by semiconductor analyst fir Read more…

AMD Clears Up Messy GPU Roadmap, Upgrades Chips Annually

June 3, 2024

In the world of AI, there's a desperate search for an alternative to Nvidia's GPUs, and AMD is stepping up to the plate. AMD detailed its updated GPU roadmap, w Read more…

Some Reasons Why Aurora Didn’t Take First Place in the Top500 List

May 15, 2024

The makers of the Aurora supercomputer, which is housed at the Argonne National Laboratory, gave some reasons why the system didn't make the top spot on the Top Read more…

Intel’s Next-gen Falcon Shores Coming Out in Late 2025 

April 30, 2024

It's a long wait for customers hanging on for Intel's next-generation GPU, Falcon Shores, which will be released in late 2025.  "Then we have a rich, a very Read more…

Leading Solution Providers

Contributors

Google Announces Sixth-generation AI Chip, a TPU Called Trillium

May 17, 2024

On Tuesday May 14th, Google announced its sixth-generation TPU (tensor processing unit) called Trillium.  The chip, essentially a TPU v6, is the company's l Read more…

Nvidia H100: Are 550,000 GPUs Enough for This Year?

August 17, 2023

The GPU Squeeze continues to place a premium on Nvidia H100 GPUs. In a recent Financial Times article, Nvidia reports that it expects to ship 550,000 of its lat Read more…

Nvidia’s New Blackwell GPU Can Train AI Models with Trillions of Parameters

March 18, 2024

Nvidia's latest and fastest GPU, codenamed Blackwell, is here and will underpin the company's AI plans this year. The chip offers performance improvements from Read more…

IonQ Plots Path to Commercial (Quantum) Advantage

July 2, 2024

IonQ, the trapped ion quantum computing specialist, delivered a progress report last week firming up 2024/25 product goals and reviewing its technology roadmap. Read more…

Choosing the Right GPU for LLM Inference and Training

December 11, 2023

Accelerating the training and inference processes of deep learning models is crucial for unleashing their true potential and NVIDIA GPUs have emerged as a game- Read more…

The NASA Black Hole Plunge

May 7, 2024

We have all thought about it. No one has done it, but now, thanks to HPC, we see what it looks like. Hold on to your feet because NASA has released videos of wh Read more…

Q&A with Nvidia’s Chief of DGX Systems on the DGX-GB200 Rack-scale System

March 27, 2024

Pictures of Nvidia's new flagship mega-server, the DGX GB200, on the GTC show floor got favorable reactions on social media for the sheer amount of computing po Read more…

MLPerf Inference 4.0 Results Showcase GenAI; Nvidia Still Dominates

March 28, 2024

There were no startling surprises in the latest MLPerf Inference benchmark (4.0) results released yesterday. Two new workloads — Llama 2 and Stable Diffusion Read more…

  • arrow
  • Click Here for More Headlines
  • arrow
HPCwire