Météo-France Fires Up Bull Supercomputer Running on ‘Broadwell’ Processors

By Linda Barney

May 9, 2016

This spring Météo-France, the national meteorological service for France and its overseas territories, turned on its ‘new’ supercomputer, a long planned upgrade delivered by Bull/Atos that doubled the performance of its predecessor (already on the Top500 list), significantly reduced power consumption, and is allowing Météo-France to increase the resolutions of its weather and climate models. It’s also one of the first such machines to use Intel’s new Xeon E5 v4 (Broadwell) processor according to Alain Beuraud, HPC Project Manager, Météo-France.

Not surprisingly, high performance computing (HPC) is a strategic element in Météo-France’s operations, critical to remaining on par with peers in the rest of Europe, the U.S., Canada, China, Korea, India, Japan, and Australia. “This is a 24/7 business,” said Beuraud, “We need reliability, accuracy and speed. There’s no room for delays in delivering our reports: after 15 minutes, the result is considered missing. Our work simply cannot be done without very high levels of computational power.”

By way of background, Météo-France provides meteorological solutions worldwide and its subsidiary MFI (Météo-France International) is recognized as a leading turnkey project integrator in the field of meteorology. Météo-France is an agency under the Ministry of Environment, Energy and Sea and has a budget of more than $480 million. It’s also a member of the European Centre for Medium-Range Weather Forecasts (ECMWF), Eumetsat, the operator of European weather satellites, and Eumetnet, the network of European meteorological services. Météo-France contributes to the World meteorological Organization (WMO) and the UN Intergovernmental Panel on Climate Change (IPCC).

In 2014, Météo-France installed two Bull systems with each containing 1008 Bull DLC B710 compute nodes, each equipped with two 12-core Intel Xeon processors E5-2600 v2 product family. “This [latest] extension of Météo-France’s supercomputer was planned from the start of the project, so that it took the Bull/Atos teams only a few weeks to replace the previous Bull blades – more than 500 of them – with new generation Bull B720 Direct Liquid Cooling blades, add several hundred blades to reach a total of 1800 compute nodes, and integrate the complete configuration within the FDR InfiniBand interconnect network,” said Damien Déclat, Head of HPC Global PreSales & Delivery at Bull/Atos.

“With the new Intel Xeon E5-2698 v4 processors, this configuration increases the available computing power by a factor of four with a sustained performance in excess of 2 PFLOPS. [Because it is] based on Bull’s energy-saving Direct Liquid Cooling technology, the FLOPS/Watt ratio is improved by a factor of 2.5. This early integration allowed Météo-France to (re)start its operations on the Bull Prolix supercomputer in early March, making it one of the first production sites worldwide equipped with Intel Xeon E5 v4 processors.”

domaine_AromeThe goal of weather and climate simulations, of course, is to provide accurate forecasts and early warning systems to aid in natural disaster prevention, better protection of property and well-being of citizens, as well as decision-making support for weather-sensitive sectors such as agriculture, marine, aviation, tourism, oil and gas, and other industries. Long-term weather analysis and climate models evaluate global climate changes to help adapt national development programs for the future.

Producing Météo-France’s weather forecasts and climate analysis is the responsibility of dozens of physicists, atmospheric chemists, hydrologists and oceanographers. Underpinning their research is a series of powerful, proprietary computer models fed by millions of data points around the world.

“Weather simulation is an extremely complex calculation that depends on numerous computer simulations of atmospheric conditions. Forecast quality depends on several factors including the number and quality of the observation points where data is collected, the methods for assimilating that data into models, the accuracy of the model itself, and the available computing power,” said Beuraud.

Météo-France’s supercomputers are used both for weather forecasting (around 80% of the resources) and the remainder for climate modeling. Noteably, the performance of supercomputers used by Météo-France from 1992 through 2016 grew from 2 GFLOPS to 5 PFLOPS. Indeed, more computational power is an ongoing need to produce ever more precise meteorological information for the French territory in Europe, not least because more precision requires increasing the number of data points to be analyzed.

With the new system, Beuraud expects to be able to extend use of Météo-France’s high-resolution model (AROME) to France’s our overseas territories. Until now, AROME 1.3 km mesh could only be run on France and Western Europe. “We will also install an “Ensemble AROME local-area forecast” to better anticipate severe events both in terms of location, and in terms of timing of the event.”

“In addition, we will also experiment with very high resolutions for the AROME model (less than 1 km) for some specific locations (around the largest French airports, for example),” stated Beuraud. For its operational and research model, the Météo-France team intensively uses many Intel tools, particularly the Fortran, C, C++ compilers, OpenMP, and Intel MPI libraries. Many developers use the Intel VTune Amplifier to optimize the performances of their application. The following figures show examples of weather modeling and precipitation mapping done using the AROME model running on the Bull supercomputer.

In the figure shown below: Example of improvement in precipitation forecast in the South of France, created using the AROME model (2.5 km mesh in 2014 -, 1.3 km mesh in 2015). Courtesy of Météo-France.

Arome_precip_29septembre_Montpellier

“As far as climate modeling is concerned,” said Beuraud, “we will run a new release of our climate model within the CMIP6 exercise (it will contribute to the sixth IPCC report, which should be published around 2020). The atmospheric part of the model – whose name is ARPEGE CLIMAT – will have a new physics, and will use higher horizontal (91 levels against 31 previously) and vertical (50 km against 150 previously) resolution. The resolution of the ocean model will also be improved (25 km instead of 100 in horizontal, 75 levels instead of 42 in vertical).”

“A very important part of our HPC developments rely on national or international cooperation. For example, our global model ARPEGE (used both for weather forecasting and climate modeling) is built in cooperation with ECMWF, this cooperation is more than 25 years old. The NEMO model that we use in our climate model comes from a large consortium, and has been developed with European cooperation between universities, research and meteorological centers.”

The adoption of the Bull/Atos supercomputers enabled Météo-France to move from vector to scalar technology and achieve much higher levels of parallel processing power for minimal total cost of ownership.

“Most importantly, the new configuration will enable us to enhance the AROME model in line with our government contract. We have moved our existing models across to the new configuration and can now develop them. But it is also helping us to deliver new and exciting services both to the French government and to our numerous clients. We are developing models for airports, for example, that will improve our ability to predict and provide useful analysis of fog conditions. It has opened up a whole new range of possibilities and cements our position at the heart of international weather and climate research,” said Beuraud.

“However, we continually need to improve the resolution of climate models and weather forecasts which requires increased computational power and supercomputers of the future will help us meet this need. As we continue our work, we need supercomputers that can allow us to process petabytes of data in an efficient manner. Météo-France will continue to utilize state-of-the-art hardware and find ways to optimize its code. We will also have to experiment with our code on new architecture, so we intend to make tests on Intel Xeon Phi coprocessors. We have our own development team and we dedicate more than 10 percent of our annual revenue to finance research and development programs to best meet our future needs.”

Author Bio:
Linda Barney is the founder and owner of Barney and Associates, a technical/marketing writing, training and web design firm in Beaverton, OR.

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!

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 XL — were added to the benchmark suite as MLPerf continues 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 power it brings to artificial intelligence.  Nvidia's DGX Read more…

Call for Participation in Workshop on Potential NSF CISE Quantum Initiative

March 26, 2024

Editor’s Note: Next month there will be a workshop to discuss what a quantum initiative led by NSF’s Computer, Information Science and Engineering (CISE) directorate could entail. The details are posted below in a Ca Read more…

Waseda U. Researchers Reports New Quantum Algorithm for Speeding Optimization

March 25, 2024

Optimization problems cover a wide range of applications and are often cited as good candidates for quantum computing. However, the execution time for constrained combinatorial optimization applications on quantum device Read more…

NVLink: Faster Interconnects and Switches to Help Relieve Data Bottlenecks

March 25, 2024

Nvidia’s new Blackwell architecture may have stolen the show this week at the GPU Technology Conference in San Jose, California. But an emerging bottleneck at the network layer threatens to make bigger and brawnier pro Read more…

Who is David Blackwell?

March 22, 2024

During GTC24, co-founder and president of NVIDIA Jensen Huang unveiled the Blackwell GPU. This GPU itself is heavily optimized for AI work, boasting 192GB of HBM3E memory as well as the the ability to train 1 trillion pa 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…

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…

NVLink: Faster Interconnects and Switches to Help Relieve Data Bottlenecks

March 25, 2024

Nvidia’s new Blackwell architecture may have stolen the show this week at the GPU Technology Conference in San Jose, California. But an emerging bottleneck at Read more…

Who is David Blackwell?

March 22, 2024

During GTC24, co-founder and president of NVIDIA Jensen Huang unveiled the Blackwell GPU. This GPU itself is heavily optimized for AI work, boasting 192GB of HB Read more…

Nvidia Looks to Accelerate GenAI Adoption with NIM

March 19, 2024

Today at the GPU Technology Conference, Nvidia launched a new offering aimed at helping customers quickly deploy their generative AI applications in a secure, s Read more…

The Generative AI Future Is Now, Nvidia’s Huang Says

March 19, 2024

We are in the early days of a transformative shift in how business gets done thanks to the advent of generative AI, according to Nvidia CEO and cofounder Jensen 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…

Nvidia Showcases Quantum Cloud, Expanding Quantum Portfolio at GTC24

March 18, 2024

Nvidia’s barrage of quantum news at GTC24 this week includes new products, signature collaborations, and a new Nvidia Quantum Cloud for quantum developers. Wh Read more…

Alibaba Shuts Down its Quantum Computing Effort

November 30, 2023

In case you missed it, China’s e-commerce giant Alibaba has shut down its quantum computing research effort. It’s not entirely clear what drove the change. 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…

Shutterstock 1285747942

AMD’s Horsepower-packed MI300X GPU Beats Nvidia’s Upcoming H200

December 7, 2023

AMD and Nvidia are locked in an AI performance battle – much like the gaming GPU performance clash the companies have waged for decades. AMD has claimed it Read more…

DoD Takes a Long View of Quantum Computing

December 19, 2023

Given the large sums tied to expensive weapon systems – think $100-million-plus per F-35 fighter – it’s easy to forget the U.S. Department of Defense is a Read more…

Synopsys Eats Ansys: Does HPC Get Indigestion?

February 8, 2024

Recently, it was announced that Synopsys is buying HPC tool developer Ansys. Started in Pittsburgh, Pa., in 1970 as Swanson Analysis Systems, Inc. (SASI) by John Swanson (and eventually renamed), Ansys serves the CAE (Computer Aided Engineering)/multiphysics engineering simulation market. 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…

Intel’s Server and PC Chip Development Will Blur After 2025

January 15, 2024

Intel's dealing with much more than chip rivals breathing down its neck; it is simultaneously integrating a bevy of new technologies such as chiplets, artificia Read more…

Baidu Exits Quantum, Closely Following Alibaba’s Earlier Move

January 5, 2024

Reuters reported this week that Baidu, China’s giant e-commerce and services provider, is exiting the quantum computing development arena. Reuters reported � Read more…

Leading Solution Providers

Contributors

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 1179408610

Google Addresses the Mysteries of Its Hypercomputer 

December 28, 2023

When Google launched its Hypercomputer earlier this month (December 2023), the first reaction was, "Say what?" It turns out that the Hypercomputer is Google's t Read more…

AMD MI3000A

How AMD May Get Across the CUDA Moat

October 5, 2023

When discussing GenAI, the term "GPU" almost always enters the conversation and the topic often moves toward performance and access. Interestingly, the word "GPU" is assumed to mean "Nvidia" products. (As an aside, the popular Nvidia hardware used in GenAI are not technically... Read more…

Shutterstock 1606064203

Meta’s Zuckerberg Puts Its AI Future in the Hands of 600,000 GPUs

January 25, 2024

In under two minutes, Meta's CEO, Mark Zuckerberg, laid out the company's AI plans, which included a plan to build an artificial intelligence system with the eq Read more…

Google Introduces ‘Hypercomputer’ to Its AI Infrastructure

December 11, 2023

Google ran out of monikers to describe its new AI system released on December 7. Supercomputer perhaps wasn't an apt description, so it settled on Hypercomputer Read more…

China Is All In on a RISC-V Future

January 8, 2024

The state of RISC-V in China was discussed in a recent report released by the Jamestown Foundation, a Washington, D.C.-based think tank. The report, entitled "E Read more…

Intel Won’t Have a Xeon Max Chip with New Emerald Rapids CPU

December 14, 2023

As expected, Intel officially announced its 5th generation Xeon server chips codenamed Emerald Rapids at an event in New York City, where the focus was really o Read more…

IBM Quantum Summit: Two New QPUs, Upgraded Qiskit, 10-year Roadmap and More

December 4, 2023

IBM kicks off its annual Quantum Summit today and will announce a broad range of advances including its much-anticipated 1121-qubit Condor QPU, a smaller 133-qu Read more…

  • arrow
  • Click Here for More Headlines
  • arrow
HPCwire