New Class of Accelerated, Environment friendly AI Techniques Mark the Subsequent Period of Supercomputing

NVIDIA immediately unveiled at SC23 the following wave of applied sciences that may raise scientific and industrial analysis facilities worldwide to new ranges of efficiency and vitality effectivity.

“NVIDIA {hardware} and software program improvements are creating a brand new class of AI supercomputers,” stated Ian Buck, vp of the corporate’s excessive efficiency computing and hyperscale information heart enterprise, in a particular tackle on the convention.

A few of the techniques will pack memory-enhanced NVIDIA Hopper accelerators, others a brand new NVIDIA Grace Hopper techniques structure. All will use the expanded parallelism to run a full stack of accelerated software program for generative AI, HPC and hybrid quantum computing.

Buck described the brand new NVIDIA HGX H200 as “the world’s main AI computing platform.”

Image of H200 GPU system
NVIDIA H200 Tensor Core GPUs pack HBM3e reminiscence to run rising generative AI fashions.

It packs as much as 141GB of HBM3e, the primary AI accelerator to make use of the ultrafast expertise. Working fashions like GPT-3, NVIDIA H200 Tensor Core GPUs present an 18x efficiency improve over prior-generation accelerators.

Amongst different generative AI benchmarks, they zip by means of 12,000 tokens per second on a Llama2-13B giant language mannequin (LLM).

Buck additionally revealed a server platform that hyperlinks 4 NVIDIA GH200 Grace Hopper Superchips on an NVIDIA NVLink interconnect. The quad configuration places in a single compute node a whopping 288 Arm Neoverse cores and 16 petaflops of AI efficiency with as much as 2.3 terabytes of high-speed reminiscence.

Image of quad GH200 server node
Server nodes based mostly on the 4 GH200 Superchips will ship 16 petaflops of AI efficiency.

Demonstrating its effectivity, one GH200 Superchip utilizing the NVIDIA TensorRT-LLM open-source library is 100x quicker than a dual-socket x86 CPU system and almost 2x extra vitality environment friendly than an X86 + H100 GPU server.

“Accelerated computing is sustainable computing,” Buck stated. “By harnessing the ability of accelerated computing and generative AI, collectively we will drive innovation throughout industries whereas decreasing our influence on the surroundings.”

NVIDIA Powers 38 of 49 New TOP500 Techniques

The newest TOP500 checklist of the world’s quickest supercomputers displays the shift towards accelerated, energy-efficient supercomputing.

Due to new techniques powered by NVIDIA H100 Tensor Core GPUs, NVIDIA now delivers greater than 2.5 exaflops of HPC efficiency throughout these world-leading techniques, up from 1.6 exaflops within the Could rankings. NVIDIA’s contribution on the highest 10 alone reaches almost an exaflop of HPC and 72 exaflops of AI efficiency.

The brand new checklist comprises the best variety of techniques ever utilizing NVIDIA applied sciences, 379 vs. 372 in Could, together with 38 of 49 new supercomputers on the checklist.

Microsoft Azure leads the newcomers with its Eagle system utilizing H100 GPUs in NDv5 situations to hit No. 3 with 561 petaflops. Mare Nostrum5 in Barcelona ranked No. 8, and NVIDIA Eos — which just lately set new AI coaching information on the MLPerf benchmarks — got here in at No. 9.

Exhibiting their vitality effectivity, NVIDIA GPUs energy 23 of the highest 30 techniques on the Green500. And so they retained the No. 1 spot with the H100 GPU-based Henri system, which delivers 65.09 gigaflops per watt for the Flatiron Institute in New York.

Gen AI Explores COVID

Exhibiting what’s attainable, the Argonne Nationwide Laboratory used NVIDIA BioNeMo, a generative AI platform for biomolecular LLMs, to develop GenSLMs, a mannequin that may generate gene sequences that carefully resemble real-world variants of the coronavirus. Utilizing NVIDIA GPUs and information from 1.5 million COVID genome sequences, it will possibly additionally quickly determine new virus variants.

The work gained the Gordon Bell particular prize final 12 months and was educated on supercomputers, together with Argonne’s Polaris system, the U.S. Division of Vitality’s Perlmutter and NVIDIA’s Selene.

It’s “simply the tip of the iceberg — the long run is brimming with potentialities, as generative AI continues to redefine the panorama of scientific exploration,” stated Kimberly Powell, vp of healthcare at NVIDIA, within the particular tackle.

Saving Time, Cash and Vitality

Utilizing the newest applied sciences, accelerated workloads can see an order-of-magnitude discount in system value and vitality used, Buck stated.

For instance, Siemens teamed with Mercedes to investigate aerodynamics and associated acoustics for its new electrical EQE automobiles. The simulations that took weeks on CPU clusters ran considerably quicker utilizing the newest NVIDIA H100 GPUs. As well as, Hopper GPUs allow them to cut back prices by 3x and cut back vitality consumption by 4x (beneath).

Chart showing the performance and energy efficiency of H100 GPUs

Switching on 200 Exaflops Starting Subsequent Yr

Scientific and industrial advances will come from each nook of the globe the place the newest techniques are being deployed.

“We already see a mixed 200 exaflops of AI on Grace Hopper supercomputers going to manufacturing 2024,” Buck stated.

They embrace the huge JUPITER supercomputer at Germany’s Jülich heart. It could actually ship 93 exaflops of efficiency for AI coaching and 1 exaflop for HPC functions, whereas consuming solely 18.2 megawatts of energy.

Chart of deployed performance of supercomputers using NVIDIA GPUs through 2024
Analysis facilities are poised to change on a tsunami of GH200 efficiency.

Primarily based on Eviden’s BullSequana XH3000 liquid-cooled system, JUPITER will use the NVIDIA quad GH200 system structure and NVIDIA Quantum-2 InfiniBand networking for local weather and climate predictions, drug discovery, hybrid quantum computing and digital twins. JUPITER quad GH200 nodes shall be configured with 864GB of high-speed reminiscence.

It’s considered one of a number of new supercomputers utilizing Grace Hopper that NVIDIA introduced at SC23.

The HPE Cray EX2500 system from Hewlett Packard Enterprise will use the quad GH200 to energy many AI supercomputers coming on-line subsequent 12 months.

For instance, HPE makes use of the quad GH200 to energy OFP-II, a sophisticated HPC system in Japan shared by the College of Tsukuba and the College of Tokyo, in addition to the DeltaAI system, which is able to triple computing capability for the U.S. Nationwide Middle for Supercomputing Purposes.

HPE can be constructing the Venado system for the Los Alamos Nationwide Laboratory, the primary GH200 to be deployed within the U.S. As well as, HPE is constructing GH200 supercomputers within the Center East, Switzerland and the U.Ok.

Grace Hopper in Texas and Past

On the Texas Superior Computing Middle (TACC), Dell Applied sciences is constructing the Vista supercomputer with NVIDIA Grace Hopper and Grace CPU Superchips.

Greater than 100 international enterprises and organizations, together with NASA Ames Analysis Middle and Whole Energies, have already bought Grace Hopper early-access techniques, Buck stated.

They be a part of beforehand introduced GH200 customers reminiscent of SoftBank and the College of Bristol, in addition to the huge Leonardo system with 14,000 NVIDIA A100 GPUs that delivers 10 exaflops of AI efficiency for Italy’s Cineca consortium.

The View From Supercomputing Facilities

Leaders from supercomputing facilities world wide shared their plans and work in progress with the newest techniques.

“We’ve been collaborating with MeteoSwiss ECMWP in addition to scientists from ETH EXCLAIM and NVIDIA’s Earth-2 undertaking to create an infrastructure that may push the envelope in all dimensions of massive information analytics and excessive scale computing,” stated Thomas Schultess, director of the Swiss Nationwide Supercomputing Centre of labor on the Alps supercomputer.

“There’s actually spectacular energy-efficiency beneficial properties throughout our stacks,” Dan Stanzione, govt director of TACC, stated of Vista.

It’s “actually the stepping stone to maneuver customers from the sorts of techniques we’ve finished prior to now to this new Grace Arm CPU and Hopper GPU tightly coupled mixture and … we’re trying to scale out by in all probability an element of 10 or 15 from what we’re deploying with Vista once we deploy Horizon in a pair years,” he stated.

Accelerating the Quantum Journey

Researchers are additionally utilizing immediately’s accelerated techniques to pioneer a path to tomorrow’s supercomputers.

In Germany, JUPITER “will revolutionize scientific analysis throughout local weather, supplies, drug discovery and quantum computing,” stated Kristel Michelson, who leads Julich’s analysis group on quantum data processing.

“JUPITER’s structure additionally permits for the seamless integration of quantum algorithms with parallel HPC algorithms, and that is obligatory for efficient quantum HPC hybrid simulations,” she stated.

CUDA Quantum Drives Progress

The particular tackle additionally confirmed how NVIDIA CUDA Quantum — a platform for programming CPUs, GPUs and quantum computer systems also called QPUs — is advancing analysis in quantum computing.

For instance, researchers at BASF, the world’s largest chemical firm, pioneered a brand new hybrid quantum-classical methodology for simulating chemical substances that may protect people towards dangerous metals. They be a part of researchers at Brookhaven Nationwide Laboratory and HPE who’re individually pushing the frontiers of science with CUDA Quantum.

NVIDIA additionally introduced a collaboration with Classiq, a developer of quantum programming instruments, to create a life sciences analysis heart on the Tel Aviv Sourasky Medical Middle, Israel’s largest instructing hospital.  The middle will use Classiq’s software program and CUDA Quantum operating on an NVIDIA DGX H100 system.

Individually, Quantum Machines will deploy the primary NVIDIA DGX Quantum, a system utilizing Grace Hopper Superchips, on the Israel Nationwide Quantum Middle that goals to drive advances throughout scientific fields. The DGX system shall be linked to a superconducting QPU by Quantware and a photonic QPU from ORCA Computing, each powered by CUDA Quantum.

Logos of NVIDIA CUDA Quantum partners

“In simply two years, our NVIDIA quantum computing platform has amassed over 120 companions [above], a testomony to its open, progressive platform,” Buck stated.

General, the work throughout many fields of discovery reveals a brand new pattern that mixes accelerated computing at information heart scale with NVIDIA’s full-stack innovation.

“Accelerated computing is paving the trail for sustainable computing with developments that present not simply superb expertise however a extra sustainable and impactful future,” he concluded.

Watch NVIDIA’s SC23 particular tackle beneath.


Leave a Reply

Your email address will not be published. Required fields are marked *