Chip manufacturing is an “splendid utility” for NVIDIA accelerated and AI computing, NVIDIA founder and CEO Jensen Huang mentioned Tuesday.
Detailing how the newest developments in computing are accelerating “the world’s most necessary trade,” Huang spoke at ITF World 2023 semiconductor convention in Antwerp, Belgium.
Huang delivered his remarks through video to a gathering of leaders from throughout the semiconductor, know-how and communications industries.
“I’m thrilled to see NVIDIA accelerated computing and AI in service of the world’s chipmaking trade,” Huang mentioned as he detailed how developments in accelerated computing, AI and semiconductor manufacturing intersect.
AI, Accelerated Computing Step Up
The exponential efficiency enhance of the CPU has been the governing dynamic of the know-how trade for practically 4 many years, Huang mentioned.
However over the previous few years CPU design has matured, he mentioned. The speed at which semiconductors change into extra highly effective and environment friendly is slowing, at the same time as demand for computing functionality soars.
“Consequently, international demand for cloud computing is inflicting information heart energy consumption to skyrocket,” Huang mentioned.
Huang mentioned that striving for web zero whereas supporting the “invaluable advantages” of extra computing energy requires a brand new strategy.
The problem is a pure match for NVIDIA, which pioneered accelerated computing, coupling the parallel processing capabilities of GPUs with CPUs.
This acceleration, in flip, sparked the AI revolution. A decade in the past, deep studying researchers reminiscent of Alex Krizhevsky, Ilya Sutskever and Geoffrey Hinton found that GPUs may very well be cost-effective supercomputers.
Since then, NVIDIA reinvented its computing stack for deep studying, opening up “multi trillion-dollar alternatives in robotics, autonomous autos and manufacturing,” Huang mentioned.
By offloading and accelerating compute-intensive algorithms, NVIDIA routinely accelerates functions by 10-100x whereas lowering energy and value by an order of magnitude, Huang defined.
Collectively, AI and accelerated computing are remodeling the know-how trade. “We’re experiencing two simultaneous platform transitions — accelerated computing and generative AI,” Huang mentioned.
AI, Accelerated Computing Come to Chip Manufacturing
Huang defined that superior chip manufacturing requires over 1,000 steps, producing options the dimensions of a biomolecule. Every step have to be practically good to yield purposeful output.
“Refined computational sciences are carried out at each stage to compute the options to be patterned and to do defect detection for in-line course of management,” Huang mentioned. “Chip manufacturing is a perfect utility for NVIDIA accelerated and AI computing.”
Huang outlined a number of examples of how NVIDIA GPUs have gotten more and more integral to chip manufacturing.
Firms like D2S, IMS Nanofabrication, and NuFlare construct masks writers — machines that create photomasks, stencils that switch patterns onto wafers — utilizing electron beams. NVIDIA GPUs speed up the computationally demanding duties of sample rendering and masks course of correction for these masks writers.
Semiconductor producer TSMC and tools suppliers KLA and Lasertech use excessive ultraviolet gentle, often called EUV, and deep ultraviolet gentle, or DUV, for masks inspection. NVIDIA GPUs play a vital function right here, too, in processing classical physics modeling and deep studying to generate artificial reference photos and detect defects.
KLA, Utilized Supplies, and Hitachi Excessive-Tech use NVIDIA GPUs of their e-beam and optical wafer inspection and overview techniques.
And in March, NVIDIA introduced that it’s working with TSMC, ASML and Synopsys to speed up computational lithography.
Computational lithography simulates Maxwell’s equations of sunshine habits passing by means of optics and interacting with photoresists, Huang defined.
Computational lithography is the biggest computational workload in chip design and manufacturing, consuming tens of billions of CPU hours yearly. Large information facilities run 24/7 to create reticles for brand spanking new chips.
Launched in March, NVIDIA cuLitho is a software program library with optimized instruments and algorithms for GPU-accelerated computational lithography.
“We’ve already accelerated the processing by 50 instances,” Huang mentioned. “Tens of hundreds of CPU servers will be changed by a couple of hundred NVIDIA DGX techniques, lowering energy and value by an order of magnitude.”
The financial savings will scale back carbon emissions or allow new algorithms to push past 2 nanometers, Huang mentioned.
What’s Subsequent?
What’s the following wave of AI? Huang described a brand new sort of AI — “embodied AI,” or clever techniques that may perceive, motive about and work together with the bodily world.
He mentioned examples embody robotics, autonomous autos and even chatbots which are smarter as a result of they perceive the bodily world.
Huang supplied his viewers a take a look at NVIDIA VIMA, a multimodal embodied AI. VIMA, Huang mentioned, can carry out duties from visible textual content prompts, reminiscent of “rearranging objects to match this scene.”
It could actually study ideas and act accordingly, reminiscent of “This can be a widget,” “That’s a factor” after which “Put this widget in that factor.” It could actually additionally study from demonstrations and keep inside specified boundaries, Huang mentioned.
VIMA runs on NVIDIA AI, and its digital twin runs in NVIDIA Omniverse, a 3D improvement and simulation platform. Huang mentioned that physics-informed AI might study to emulate physics and make predictions that obey bodily legal guidelines.
Researchers are constructing techniques that mesh info from actual and digital worlds on an unlimited scale.
NVIDIA is constructing a digital twin of our planet, referred to as Earth-2, which is able to first predict the climate, then long-range climate, and ultimately local weather. NVIDIA’s Earth-2 group has created FourCastNet, a physics-AI mannequin that emulates international climate patterns 50-100,000x sooner.
FourCastNet runs on NVIDIA AI, and the Earth-2 digital twin is inbuilt NVIDIA Omniverse.
Such techniques promise to deal with the best problem of our time, reminiscent of the necessity for reasonable, clear vitality.
For instance, researchers on the U.Ok.’s Atomic Power Authority and the College of Manchester are making a digital twin of their fusion reactor, utilizing physics-AI to emulate plasma physics and robotics to manage the reactions and maintain the burning plasma.
Huang mentioned scientists might discover hypotheses by testing them within the digital twin earlier than activating the bodily reactor, bettering vitality yield, predictive upkeep and lowering downtime. “The reactor plasma physics-AI runs on NVIDIA AI, and its digital twin runs in NVIDIA Omniverse,“ Huang mentioned.
Such techniques maintain promise for additional developments within the semiconductor trade. “I stay up for physics-AI, robotics and Omniverse-based digital twins serving to to advance the way forward for chip manufacturing,” Huang mentioned.
