Now Reading
Nvidia Speeds Key Chipmaking Computation by 40x

Nvidia Speeds Key Chipmaking Computation by 40x

2023-03-25 19:17:45

Nvidia says it has discovered a method to pace up a computation-limited step within the chipmaking course of in order that it occurs 40 instances as quick as as we speak’s normal. Known as inverse lithography, it’s a key software that permits chipmakers to print nanometer-scale options utilizing mild with an extended wavelength than the dimensions of these options. Inverse lithography’s use has been restricted by the large measurement of the wanted computation. Nvidia’s reply, cuLitho, is a set of algorithms designed to be used with GPUs, turns what has been two weeks of labor into an in a single day job.

The know-how “will permit fabs to extend throughput, cut back their carbon footprint, and set the foundations for 2-nanometer [manufacturing processes] and past,” stated Nvidia CEO Jensen Huang at the Nvidia GTC developer conference on Tuesday.

Main logic-chip foundry Taiwan Semiconductor Manufacturing Co. (TSMC) can be qualifying cuLitho’s use in manufacturing beginning in June, stated Huang. Design automation software program agency Synopsys plans to combine software program, too, and lithography tools maker ASML plans to assist cuLitho in its merchandise as effectively.

Photolithography is mainly step one within the chipmaking course of. It entails bouncing mild off of a sample referred to as a photomask to venture the types of transistor constructions and interconnects onto the wafer. (Extra mature know-how makes use of transmissive photomasks as a substitute of reflective ones, however the concept is similar.) It takes 98 photomasks to make an H100 GPU, Jensen stated. The options projected from the photomask are smaller than the wavelength of sunshine used—193 nanometers for the comparatively massive options and 13.5 nm for the finer bits. So, with out assistance from tips and design guidelines—collectively referred to as optical proximity correction—you’d get solely a blurry mess projected onto the wafer. However with optical proximity correction, the designs on the photomask solely vaguely resemble the sample of sunshine on the chip.

With the necessity for finer and finer options, the corrected shapes on the photomask have turn out to be an increasing number of elaborate and tough to provide you with. It will be significantly better to start out with the sample you need on the wafer after which calculate what sample on the photomask would produce them. Such a scheme known as inverse lithography. Easy because it sounds, it’s fairly tough to compute, typically taking weeks to compile.

In truth, it’s such a slog that it’s typically reserved to be used on only some essential layers of modern chips or simply significantly thorny bits of them, based on knowledge from the E-Beam Initiative, which periodically surveys the business.

Images on a timeline start as a simple cross and become more complicated as time goes on.As chipmaking required finer and finer options, engineers needed to produce an increasing number of complicated designs to venture these options onto the silicon. Inverse lithography (ILT) is the newest improvement.Nvidia

The lengthy computation time for lithography slows the event and enchancment of chip know-how Even a change to the thickness of a cloth can result in the necessity for a brand new set of photomasks, notes Vivek K. Singh, vp within the superior know-how group engaged on silicon manufacturing at Nvidia. Computing masks “has been a protracted pole in chip improvement,” he says. “If inverse lithography know-how have been sped up 40 instances, would extra firms apply it to extra layers? Absolutely.”

A part of the computation is a picture downside that’s a pure match for GPUs, says Singh. However at most, that may solely minimize the computation time in half. The remainder shouldn’t be really easy to make parallel. However over the previous 4 years, with improvement companions together with TSMC, Nvidia engineers have provide you with a set of algorithms for making the remaining work parallel and have packaged it as a software program library to be used with GPUs.

In keeping with Nvidia, utilizing the cuLitho lets 500 Nvidia DGX H100 computer systems do the work of 40,000 CPU programs. It could produce three to 5 instances as many photomasks per day, drawing solely 5 megawatts as a substitute of 35 MW.

See Also

What’s extra, the know-how could ship higher outcomes, based on Singh. CuLitho produces in any other case hard-to-calculate curvy polygons on the masks, which leads to a better depth of focus for the sample solid onto the wafer. That depth of focus ought to result in much less variation throughout the wafer and due to this fact a better yield of working chips per wafer, he says. In future, it additionally might imply fewer photomasks are wanted; with inverse lithography, what should now be achieved with a double sample may work with just one.

Nvidia shouldn’t be the primary to look to GPUs to speed up inverse lithography know-how. Silicon Valley-based D2S introduced a GPU-based computer customized constructed for the issue in 2019. IEEE Spectrum reached out to D2S for remark, however the firm didn’t reply earlier than press time.

From Your Website Articles

Associated Articles Across the Internet

Source Link

What's Your Reaction?
In Love
Not Sure
View Comments (0)

Leave a Reply

Your email address will not be published.

2022 Blinking Robots.
WordPress by Doejo

Scroll To Top