Russian scientists suggest a PC to solve complex problems tens of times faster than with massive supercomputers
A group of physicists in Russia has learned to use a personal computer for calculations of complex equations of quantum mechanics, usually solved with help of supercomputers. This PC does the job much faster.
A group of physicists from the Skobeltsyn Institute of Nuclear Physics, the Lomonosov Moscow State University, has learned to use a personal computer for calculations of complex equations of quantum mechanics, usually solved with help of supercomputers. This PC does the job much faster. An article about the results of the work has been published in the journal Computer Physics Communications.
Senior researchers Vladimir Pomerantcev and Olga Rubtsova, working under the guidance of Professor Vladimir Kukulin (SINP MSU), were able to use on an ordinary desktop PC with GPU to solve complicated integral equations of quantum mechanics — previously solved only with the powerful, expensive supercomputers. According to Vladimir Kukulin, the personal computer does the job much faster: in 15 minutes it is doing the work requiring normally 2-3 days of the supercomputer time.
The equations in question were formulated in the ’60s by the Russian mathematician Ludwig Faddeev. The equations describe the scattering of a few quantum particles, i.e., represent a quantum mechanical analog of the Newtonian theory of the three body systems. As the result, the whole field of quantum mechanics called “physics of few-body systems” appeared soon after this.
This area poses a great interest to scientists engaged in quantum mechanics, nuclear and atomic physics and the theory of scattering. For several decades after the pioneering work of Faddeev one of their main purposes was to learn to solve these complicated equations. However, due to the incredible complexity of the calculations in the case of fully realistic interactions between a system’s particles stood out of the researchers’ reach for a long time, until the supercomputers appeared.
The situation changed dramatically after the group of SINP decided to use one of the new Nvidia GPs designed for use in game consoles on their personal computer. According to one of the authors Vladimir Kukulin, Head of Laboratory of Nuclear Theory, the processor was not the most expensive, of those that you can buy for $300-500.
The main problem in solving the scattering equations of multiple quantum particles was the calculation of the integral kernel — a huge two-dimensional table, consisting of tens or hundreds of thousands of rows and columns, with each element of such a huge matrix being the result of extremely complex calculations. But this table appeared to look like a monitor screen with tens of billions of pixels, and with a good GPU it was quite possible to calculate all of these. Using the software developed in Nvidia and having written their own programs, the researchers split their calculations on the many thousands of streams and were able to solve the problem brilliantly.
“We reached the speed we couldn’t even dream of,” Vladimir Kukulin said.
Learn more: No need for supercomputers
The Latest on: PCs vs Supercomputers
[google_news title=”” keyword=”PCs vs Supercomputers” num_posts=”10″ blurb_length=”0″ show_thumb=”left”]
via Google News
The Latest on: PCs vs Supercomputers
- Students use 16 x NVIDIA Jetson Nano modules to make desktop supercomputeron November 17, 2022 at 1:58 pm
The desktop supercomputer is powered by 16 x NVIDIA ... Working in IT retail for 10 years gave him great experience with custom-built PCs. His addiction to GPU tech is unwavering.
- Supercomputers Market Share 2022 Global Business Industry Revenue, Demand and Applications Market Research Report to 2026on November 17, 2022 at 11:17 am
Nov 17, 2022 (The Expresswire) -- Global Supercomputers Market research report presents an In-Depth detailed overview of the current market situation and forecast till 2026. The study perhaps a ...
- Nvidia Is Building an AI Supercomputer With Microsofton November 17, 2022 at 9:00 am
I've been working at PCMag since November 2016, covering all areas of technology and video game news. Before that I spent nearly 15 years working at Geek.com as a writer and editor. I also spent ...
- Nvidia and Microsoft team up to build ‘massive’ AI supercomputeron November 17, 2022 at 3:44 am
Nvidia has announced a “multi-year collaboration” with Microsoft to build “one of the most powerful AI supercomputers in the world,” designed to handle the huge computing workloads needed ...
- HPE launches supercomputers for the enterpriseon November 16, 2022 at 5:08 am
You can now get a scaled-down and less expensive supercomputer that's based on the same architecture of the machine that broke the exaFLOP barrier. Supercomputers are super expensive, but Hewlett ...
- This giant supercomputer dominates the rankings, and catching it will take a lot of workon November 16, 2022 at 3:02 am
Frontier, the HPE Cray exa-scale supercomputer operated by the Department of Energy (DoE) at Oak Ridge National Laboratory (ORNL), remains by far the fastest supercomputer on the planet ...
- A stack of Apple laptops could work as a powerful supercomputeron November 16, 2022 at 1:42 am
A cluster of standard Apple laptops could work as a supercomputer, enabling people to do computationally demanding tasks in a cheaper way. Researchers who need powerful computers to run their ...
- AMD is the supercomputer kingon November 15, 2022 at 3:21 am
AMD continues to dominate the Top 500 list of the fastest supercomputers in the world. The latest list shows that the AMD-powered exascale-class Frontier still ranks as the fastest system in the ...
- New Flatiron Institute Supercomputer the Most Power-Efficient Ever Builton November 14, 2022 at 7:59 pm
The revised list, which ranks supercomputers based on the number of floating-point operations per second (or ‘FLOPS’) per watt of power, clocks the new supercomputer at 65.091 billion FLOPS ...
- Hungry for AI? New supercomputer contains 16 dinner-plate-size chipson November 14, 2022 at 2:03 pm
On Monday, Cerebras Systems unveiled its 13.5 million core Andromeda AI supercomputer for deep learning, reports Reuters. According Cerebras, Andromeda delivers over one 1 exaflop (1 quintillion ...
via Bing News