0
Intel-Powered Aurora Becomes The Fastest Supercomputer For AI, Finally Breaks The Exascale Barrier

Intel-Powered Aurora Becomes The Fastest Supercomputer For AI, Finally Breaks The Exascale Barrier

The Aurora supercomputer has finally broken the exascale barrier and achieved the fastest AI performance with its Intel Ponte Vecchio hardware.

The Intel-powered Aurora supercomputer achieves 1.01 exaflops in compute and 10.6 AI exaflops as the system reaches full operational capacity.

Deployed at Argonne National Laboratory and developed in collaboration with HPE (Hewlett Packard Enterprise), Aurora Supercomputer It was expected to be one of the top performers in the HPC and AI segment. Powered by Intel's Xeon CPU Max and Data Center GPU Max series, the platform was competitive. AMD managed to be the first to hit the exascale barrier.. Meanwhile, despite being relaunched in 2019, the Aurora supercomputer barely managed to meet its targets but today, the system has reached 87% operational capacity or a total of 9234 nodes.

At ISC High Performance 2024, Intel, in collaboration with Argonne National Laboratory and Hewlett Packard Enterprise (HPE), announced that the Aurora supercomputer has broken the exascale barrier at 1.012 exaflops and is the world's fastest AI system Open Science is dedicated to AI, achieving 10.6 AI exaflops. Intel will also detail the critical role of open ecosystems in driving AI-accelerated high-performance computing (HPC).

By Intel

In terms of specifications, the Aurora supercomputer is built using 166 racks with 10,624 blades, 21,248 Intel Xeon CPU Max chips (4th Gen Sapphire Rapids), and 63,744 Intel Data Center GPU Max series units (Ponte Vecchio). It is based on HPE Slingshot fabric for interconnect and uses 84,992 endpoints.

Image source: ISC24

In terms of performance metrics, the Aurora supercomputer managed to come second in the HPL LINPACK benchmark but managed to break the exascale barrier by utilizing only 87% of the total node capacity (9234 nodes vs. 10,624) at 1.012 exaflops. . Ranked third in the HPCG test with 5612 TFLOPs/sec using only 39% of the system.

Using the Xe core architecture and its multiple AI hardware blocks, the Aurora supercomputer now manages to rank 1st in AI performance charts with a total rated performance of 10.6 AI Exaflops. Performance was measured using the LINPACK-mixed precision (HPL-MxP) benchmark.

Image source: ISC24

what's next: The new supercomputers being deployed with Intel Xeon CPU Max Series and Intel Data Center GPU Max Series technologies underline Intel's goal to advance HPC and AI. Systems include the Euro-Mediterranean Center on Climate Change (CMCC) Cassandra to accelerate climate change modelling. Italian National Agency for New Technologies, Energy and Sustainable Economic Development (ENEA) CRESCO 8 for achievements in fusion energy; The Texas Advanced Computing Center (TACC), which is in full production to enable data analysis in biology, supersonic turbulence flow and atomistic simulations on a wide range of materials. as well as the United Kingdom Atomic Energy Authority (UKAEA) to address memory-related issues that underpin the design of future fusion power plants.

The result of the mixed-precision AI benchmark will be the basis for Intel's next-generation GPUs for AI and HPC. Code name Falcon Shores. Falcon Shores will leverage next-generation Intel Xe architecture with the best of Intel Gaudi. This integration enables a unified programming interface.

Preliminary performance results on Intel Xeon with 6 P-cores And multiplexer combined rank (MCR) memory at 8800 MT/s provides up to 2.3x performance improvement for real-world HPC applications, such as the Nucleus for European Modeling of the Ocean (NEMO), when compared to the previous generation. , establishes a strong foundation. As the preferred host CPU choice for HPC solutions.

Share this story.

Facebook

Twitter

About the Author

Leave a Reply