Intel has announced that the Aurora supercomputer has broken the exascale barrier at 1.012 exaflops, becoming 'the world's fastest AI system dedicated to AI for open science'.

"Whether it's understanding climate models or unlocking the mysteries of the universe, supercomputers serve as a compass to guide us towards solving truly difficult scientific challenges that could improve humanity," emphasized Ogi Brkic, Intel vice president and general manager of Data Center AI Solutions.

In detail, the Aurora supercomputer is an extended system featuring 166 racks, 10624 compute blades, 21248 Intel® Xeon® CPU Max Series processors and 63744 Intel® Data Center GPU Max Series units.

The Aurora supercomputer came second in the LINPACK high-performance benchmark (HPL), but broke the exascale barrier at 1,012 exaflops, using 9234 nodes - just 87% of the system.

The Aurora supercomputer also secured third place on the high-performance conjugate gradient (HPCG) benchmark at 5612 teraflops per second (TF/s), using 39% of the machine.

Copyright (c) 2024 All rights reserved.