{"id":2893,"date":"2023-02-28T17:39:48","date_gmt":"2023-02-28T17:39:48","guid":{"rendered":"https:\/\/salarydistribution.com\/machine-learning\/2023\/02\/28\/fusion-reaction-how-ai-hpc-are-energizing-science\/"},"modified":"2023-02-28T17:39:48","modified_gmt":"2023-02-28T17:39:48","slug":"fusion-reaction-how-ai-hpc-are-energizing-science","status":"publish","type":"post","link":"https:\/\/salarydistribution.com\/machine-learning\/2023\/02\/28\/fusion-reaction-how-ai-hpc-are-energizing-science\/","title":{"rendered":"Fusion Reaction: How AI, HPC Are Energizing Science"},"content":{"rendered":"
\n

Brian Spears says his children will enjoy a more sustainable planet, thanks in part to AI and high performance computing (HPC) simulations.<\/p>\n

\u201cI believe I\u2019ll see fusion energy in my lifetime, and I\u2019m confident my daughters will see a fusion-powered world,\u201d said the 45-year-old principal investigator at Lawrence Livermore National Laboratory who helped demonstrate the physics of the clean and abundant power source, making headlines worldwide.<\/p>\n

Results from the experiment hit Spears\u2019 inbox at 5:30 a.m. on Dec. 5 last year.<\/p>\n

\u201cI had to rub my eyes to make sure I wasn\u2019t misreading the numbers,\u201d he recalled.<\/p>\n

A Nuclear Family\u00a0\u00a0<\/b><\/h2>\n

Once he assured himself, he scurried downstairs to share the news with his wife, a chemical engineer at the lab who\u2019s pioneering ways to 3D print glass, and also once worked on the fusion program.<\/p>\n

\"LLNL<\/a>
Brian Spears<\/figcaption><\/figure>\n

\u201cOne of my friends described us as a Star Trek<\/i> household \u2014 I work on the warp core and she works on the replicator,\u201d he quipped.<\/p>\n

In a tweet storm after the lab formally announced the news, Spears shared his excitement with the world.<\/p>\n

\u201cExhausted by an amazing day \u2026 Daughters sending me screenshots with breaking news about Mom and Dad\u2019s work \u2026 Being a part of something amazing for humanity.\u201d<\/p>\n

In another tweet<\/a>, he shared the technical details.<\/p>\n

\u201cUsed two million joules of laser energy to crush a capsule 100x smoother than a mirror. It imploded to half the thickness of a hair. For 100 trillionths of a second, we produced ten petawatts of power. It was the brightest thing in the solar system.\u201d<\/p>\n

AI Helps Call the Shots<\/b><\/h2>\n

A week before the experiment, Spears\u2019 team analyzed its precision HPC design, then predicted the result with AI. Two atoms would fuse into one, releasing energy in a process simply called ignition.<\/p>\n

It was the most exciting of thousands of AI predictions in what\u2019s become the two-step dance of modern science. Teams design experiments in HPC simulations, then use data from the actual results to train AI models that refine the next simulation.<\/p>\n

AI uncovers details about the experiments hard for humans to see. For example, it tracked the impact of minute imperfections in the imploding capsule researchers blasted with 192 lasers to achieve fusion.<\/p>\n

\"LLNL<\/a>
A look inside the fusion experiment. Graphic courtesy of Lawrence Livermore National Laboratory.<\/figcaption><\/figure>\n

\u201cYou need AI to understand the complete picture,\u201d Spears said.<\/p>\n

It\u2019s a big canvas, filled with math describing the complex details of atomic physics.<\/p>\n

A single experiment can require hundreds of thousands of relatively small simulations. Each takes a half day on a single node of a supercomputer.<\/p>\n

The largest 3D simulations \u2014 called the kitchen sinks \u2014 consume about half of Sierra, the world\u2019s sixth fastest HPC system, packing 17,280 NVIDIA GPUs.<\/p>\n

Edge AI Guides Experiments<\/b><\/h2>\n

AI also helps scientists create self-driving experiments. Neural networks can make split-second decisions about which way to take an experiment based on results they process in real time.<\/p>\n

For example, Spears, his colleagues and NVIDIA collaborated on an AI-guided experiment last year that fired lasers up to three times a second. It created the kind of proton beams that could someday treat a cancer patient.<\/p>\n

\u201cIn the course of a day, you can get the kind of bright beam that may have taken you months or years of human-designed experiments,\u201d Spears said. \u201cThis approach of AI at the edge will save orders of magnitude of time for our subject-matter experts.\u201d<\/p>\n

Directing lasers fired many times a second will also be a key job inside tomorrow\u2019s nuclear fusion reactors.<\/p>\n

Navigating the Data Deluge<\/b><\/h2>\n

AI\u2019s impacts will be felt broadly across both scientific and industrial fields, Spears believes.<\/p>\n

\u201cOver the last decade we\u2019ve produced more simulation and experimental data than we\u2019re trained to deal with,\u201d he said.<\/p>\n

That deluge, once a burden for scientists, is now fuel for machine learning.<\/p>\n

\u201cAI is putting scientists back in the driver seat so we can move much more quickly,\u201d he said.<\/p>\n

\"Brian<\/a>
Spears explained the ignition result in an interview<\/a> (starting 8:19) with Government Matters.<\/figcaption><\/figure>\n

Spears also directs an AI initiative<\/a> at the lab that depends on collaborations with companies including NVIDIA.<\/p>\n

\u201cNVIDIA helps us look over the horizon, so we can take the next step in using AI for science,\u201d he said<\/p>\n

A Brighter Future<\/b><\/h2>\n

It\u2019s hard work with huge impacts, like leaving a more resilient planet for the next generation.<\/p>\n

Asked whether his two daughters plan a career in science, Spears beams. They\u2019re both competitive swimmers who play jazz trumpet with interests in everything from bioengineering to art.<\/p>\n

\u201cAs we say in science, they\u2019re four pi, they cover the whole sky,\u201d he said.<\/p>\n<\/div>\n","protected":false},"excerpt":{"rendered":"

https:\/\/blogs.nvidia.com\/blog\/2023\/02\/28\/ai-hpc-nuclear-fusion\/<\/p>\n","protected":false},"author":0,"featured_media":2894,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":[],"categories":[3],"tags":[],"_links":{"self":[{"href":"https:\/\/salarydistribution.com\/machine-learning\/wp-json\/wp\/v2\/posts\/2893"}],"collection":[{"href":"https:\/\/salarydistribution.com\/machine-learning\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/salarydistribution.com\/machine-learning\/wp-json\/wp\/v2\/types\/post"}],"replies":[{"embeddable":true,"href":"https:\/\/salarydistribution.com\/machine-learning\/wp-json\/wp\/v2\/comments?post=2893"}],"version-history":[{"count":0,"href":"https:\/\/salarydistribution.com\/machine-learning\/wp-json\/wp\/v2\/posts\/2893\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/salarydistribution.com\/machine-learning\/wp-json\/wp\/v2\/media\/2894"}],"wp:attachment":[{"href":"https:\/\/salarydistribution.com\/machine-learning\/wp-json\/wp\/v2\/media?parent=2893"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/salarydistribution.com\/machine-learning\/wp-json\/wp\/v2\/categories?post=2893"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/salarydistribution.com\/machine-learning\/wp-json\/wp\/v2\/tags?post=2893"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}