Tag:

HPC

El Capitan

AMD supercomputers take gold and silver in latest Top500 as Chinese HPC remains shrouded in secrecy

by admin June 11, 2025

Top500.org on Tuesday released its 65th list of the world’s most powerful supercomputers, revealing the dominance of American AMD-based systems amid a lack of new entries from China, as well as the performance of supercomputers used by AI giants such as xAI and OpenAI.

The latest update to the global high-performance computing rankings places the AMD Instinct MI300A-based El Capitan at the forefront with Rmax performance of 1.7 FP64 ExaFLOPS, followed by the AMD-powered Frontier (1.353 ExaFLOPS) and Intel-based Aurora (1.012 ExaFLOPS) as the top three exascale-class systems, all of which are operated by U.S. Department of Energy laboratories.

The world’s fourth-most most powerful supercomputer — Germany’s Jupiter Booster, based on Nvidia’s GH200 platform — is a new entrant offering Rmax performance of 0.793 ExaFLOPS. Rounding out the top five, Microsoft’s Eagle achieved 0.561 ExaFLOPS.

You may like

El Capitan, located at Lawrence Livermore National Laboratory in California, achieves 1.742 ExaFLOPS on the HPL benchmark using AMD’s fourth-generation EPYC processors and Instinct MI300A accelerators, all within an HPE Cray EX255a framework. It comprises over 11 million cores and uses the HPE Slingshot interconnect.

El Capitan also topped the companion High Performance Conjugate Gradients (HPCG) performance list with 17.1 PetaFLOPS and in the HPL-MxP mixed-precision benchmark with 16.7 ExaFLOPS. As for energy efficiency, it stands at 58.9 GigaFLOPS per watt.

Frontier, the second-place system in HPL, is installed at Oak Ridge National Laboratory in Tennessee and achieved 1.353 ExaFLOPS. It uses AMD’s third-generation EPYC CPUs, Radeon Instinct 250X accelerators, and HPE Cray EX235a infrastructure with the same Slingshot interconnect. It operates with over 8.6 million cores.

Frontier ranks third in the HPCG benchmark (behind Japan’s Fugaku) with 14.05 petaflops. Frontier also took third place in the HPL-MxP benchmark with 11.4 ExaFLOPS. Its energy efficiency works out to 54.98 GigaFLOPS per watt.

Aurora is housed at Argonne National Laboratory in Illinois and recorded 1.012 ExaFLOPS on the HPL benchmark for the bronze medal in HPL. Built on Intel Xeon CPU Max and Data Center GPU Max (aka Ponte Vecchio) components, it relies on an HPE Cray EX installation using the Slingshot interconnect. It secured the second spot in the HPL-MxP benchmark with 11.6 ExaFLOPS, but it’s in fourth place on the HPCG benchmark at 5.6 PFLOPS.

JUPITER Booster is Europe’s first system in this performance class. The machine is installed at the Jülich Supercomputing Centre in Germany and reaches its fourth-place spot with a preliminary 793.4 PFLOPS in HPL. JUPITER Booster uses Nvidia Grace Hopper hardware on Eviden’s BullSequana XH3000 platform with direct liquid cooling and HP’s Slingshot networking. It is currently being brought online.

Microsoft Azure’s Eagle system holds the fifth position in HPL with 561 PFLOPS using Xeon Platinum 8480C processors and Nvidia H100 GPUs.

Although CPUs from AMD power two of the world’s highest-performing processors and five out of Top 10 machines, a detailed analysis of the June 2025 Top 500 list reveals that Intel processors are used in 294 of the 500 systems. AMD follows with 173 supercomputers.

Nvidia systems appear in 13 entries, showing early traction for the Arm-compatible Grace Hopper architecture. Nine systems are based on other Arm processors, such as Fujitsu’s A64FX. An additional six systems use other processor types, including IBM Power9 and China’s Sunway architecture.

The U.S. extended its numerical lead in total systems in the 65th list, while China continued its downward trend as it no longer submits results of its latest systems to Top500.org. As a result, there are 175 American systems, 47 Chinese supercomputers, and 41 German machines in the Top 500 list.

On the energy efficiency front, Germany’s JEDI system leads with 72.73 GigaFLOPS per watt, followed by France’s ROMEO-2025 at 70.91 and Adastra 2 at 69.1. All three use BullSequana XH3000 infrastructure. El Capitan and Frontier ranked 26th and below on energy, reflecting a different balance between performance and efficiency.

Follow Tom’s Hardware on Google News to get our up-to-date news, analysis, and reviews in your feeds. Make sure to click the Follow button.

June 11, 2025 0 comments

Nord Quantique quantum computing

Quantum startup claims its 20-square-meter machine will crush HPC giants and rewrite the future of data centers forever

by admin June 2, 2025

Nord Quantique promises quantum power without the bulk or energy drain
Traditional HPC may fall if Nord’s speed and energy claims prove real
Cracking RSA-830 in an hour could transform cybersecurity forever

A quantum computing startup has announced plans to develop a utility-scale quantum computer with more than 1,000 logical qubits by 2031.

Nord Quantique has set an ambitious target which, if achieved, could signal a seismic shift in high-performance computing (HPC).

The company claims its machines are smaller and would offer far greater efficiency in both speed and energy consumption, thereby making traditional HPC systems obsolete.

You may like

Advancing error correction through multimode encoding

Nord Quantique uses “multimode encoding” via a technique known as the Tesseract code, and this allows each physical cavity in the system to represent more than one quantum mode, effectively increasing redundancy and resilience without adding complexity or size.

“Multimode encoding allows us to build quantum computers with excellent error correction capabilities, but without the impediment of all those physical qubits,” explained Julien Camirand Lemyre, CEO of Nord Quantique.

“Beyond their smaller and more practical size, our machines will also consume a fraction of the energy, which makes them appealing for instance to HPC centers where energy costs are top of mind.”

Nord’s machines would occupy a mere 20 square meters, making them highly suitable for data center integration.

Compared to 1,000–20,000 m² needed by competing platforms, this portability further strengthens its case.

“These smaller systems are also simpler to develop to utility-scale due to their size and lower requirements for cryogenics and control electronics,” the company added.

The implication here is significant: better error correction without scaling physical infrastructure, a central bottleneck in the quantum race.

In a technical demonstration, Nord’s system exhibited excellent stability over 32 error correction cycles with no measurable decay in quantum information.

“Their approach of encoding logical qubits in multimode Tesseract states is a very effective method of addressing error correction and I am impressed with these results,” said Yvonne Gao, Assistant Professor at the National University of Singapore.

“They are an important step forward on the industry’s journey toward utility-scale quantum computing.”

Such endorsements lend credibility, but independent validation and repeatability remain critical for long-term trust.

Nord Quantique claims its system could solve RSA-830, a representative cryptographic challenge, in just one hour using 120 kWh of energy at 1 MHz speed, slashing the energy need by 99%.

In contrast, traditional HPC systems would require approximately 280,000 kWh over nine days. Other quantum modalities, such as superconducting, photonic, cold atoms, and ion traps, fall short in either speed or efficiency.

For instance, cold atoms might consume only 20 kW, but solving the same problem would take six months.

That said, there remains a need for caution. Post-selection – used in Nord’s error correction demonstrations, required discarding 12.6% of data per round. While this helped show stability, it introduces questions about real-world consistency.

In quantum computing, the leap from laboratory breakthrough to practical deployment can be vast; thus, the claims on energy reduction and system miniaturization, though striking, need independent real-world verification.

You might also like

June 2, 2025 0 comments