The Supercomputer Set to Shake the Science We Know

The claimed world’s fastest artificial intelligent supercomputer has just launched, named Perlmutter after astrophysicist Saul Perlmutter, at the National Energy Research Scientific Computing Center (NERSC) in California with a challenging first task by applying its artificial intelligence to create a 3D map of the universe. Partnered with Nvidia which brings to the table 6,159 of its Nvidia A100 Tensor Core graphics processing units, capable of rendering quintillion floating-point operations per second, according to Nvidia specialised for artificial intelligence, data analytics, and HPC (High-performance computing) perfect for the mighty task Perlmutter is set to execute.

Perlmutter’s A100 Tensor Cores are uniquely suited to help with this, as they’re able to accelerate both the double-precision floating-point math for simulations and the mixed-precision calculations required for deep learning.
— Dion Harris

The reasoning behind building a 3D map of the universe is to enable scientists to learn more about dark energy an unknown form of energy that is said to be the cause of the accelerating rate of expansion of the universe. But why is it so important? Well, more knowledge of dark energy could challenge our understanding of gravity, perhaps into a theory that no one has even thought about, leading scientists to also question the Big Bang, from the birth of the universe, how it's aged, and how it will be ultimately destroyed. It's clear that dark energy is something that needs to be understood as it leads to question science we have all grown up learning about.

Perlmutter can also benefit material science laying the way for advances in batteries and biofuels. Also, applications like the Quantum Espresso (an integrated suite of Open-Source computer codes for electronic-structure calculations and materials modelling at the nanoscale) can take advantage of Perlmutter’s traditional simulation and machine learning capabilities, which will enables scientists to study more atoms over a longer time period.

In the past it was impossible to do fully atomistic simulations of big systems like battery interfaces, but now scientists plan to use Perlmutter to do just that
— Brandon Cook

Perlmutter is running as you read this and we should all be very excited for the results put forward by the processing powerhouse.

Jamilla Kone