Tech

How Nvidia Is Using AI To Make Even Faster And Better GPUs

The first area is mapping the voltage drop to determine where the current is being drawn on the Nvidia GPU. In a presentation, Dally explained that it would take 3 hours to run it manually in CAD, but the same job could be done in about 18 minutes with an AI-enabled GPU setup. The second area relates to parasite testing to see how circuit design performs, a common process adopted by AI. In the third area, AI-powered GPUs test different layouts of the chip to determine the most congested design format. Finally, GPUs are also used to create new designs. Nvidia’s NVCell technology uses reinforcement learning to act as an automatic default cell layout generator. Dally explained that whenever technology advances, such as moving a manufacturing process from a 7nm to a 5nm node, thousands of these cells must be redesigned to:A very complex set of design rules.” Nvidia’s NVCell is capable of rebuilding 92% of its cell library as if it were seemingly error-free.

Of note, transplanting a cell library with a new technique would require a group of 10 people to work for at least a year. Dally explained that with the help of some powerful Nvidia GPUs, you can achieve the same results in a matter of days. Of course, all these areas still require human intervention as they are forward-looking. However, AI-enabled Nvidia GPUs will help the company save a lot of time and produce better-designed chips. in addition to the GPU nvidia You could also get into the CPU manufacturing business soon, but that’s another story.

Source: NVIDIA


More information

How Nvidia Is Using AI To Make Even Faster And Better GPUs

The first area involves mapping voltage drops to determine where the power is used in an Nvidia GPU. Dally explained during the presentation that running it manually on CAD would take three hours, but with the help of an AI-powered GPU setup, the same can be accomplished in around 18 minutes. The second area involves testing parasitics to check how a circuit design would perform, which is a frequentative process that AI handles. In the third area, an AI-powered GPU tests different layouts of the chips to determine the least congested design format. And lastly, GPUs are used to create new designs as well. Nvidia’s NVCell technology uses Reinforcement Learning to work as an automatic standard cell layout generator. Dally explained that whenever technology evolves, like the transition of the fabrication process from 7nm to 5nm nodes, thousands of these cells have to be redesigned using “a very complex set of design rules.” Nvidia’s NVCell can recreate 92 percent of the cell library with seemingly no error.
For reference, it would require a group of 10 people to work for over a year to port a new technology cell library. The same can be accomplished with the help of a few powerful Nvidia GPUs in a matter of days, explained Dally. Of course, human intervention is still required in all these areas, as futuristic as it sounds. However, the AI-enabled Nvidia GPUs help the company massively save time and make a better-designed chip. In addition to GPUs, Nvidia might soon dip its toes in the CPU manufacturing business as well, but that’s a story for another time.
Source: Nvidia

#Nvidia #Faster #GPUs

How Nvidia Is Using AI To Make Even Faster And Better GPUs

The first area involves mapping voltage drops to determine where the power is used in an Nvidia GPU. Dally explained during the presentation that running it manually on CAD would take three hours, but with the help of an AI-powered GPU setup, the same can be accomplished in around 18 minutes. The second area involves testing parasitics to check how a circuit design would perform, which is a frequentative process that AI handles. In the third area, an AI-powered GPU tests different layouts of the chips to determine the least congested design format. And lastly, GPUs are used to create new designs as well. Nvidia’s NVCell technology uses Reinforcement Learning to work as an automatic standard cell layout generator. Dally explained that whenever technology evolves, like the transition of the fabrication process from 7nm to 5nm nodes, thousands of these cells have to be redesigned using “a very complex set of design rules.” Nvidia’s NVCell can recreate 92 percent of the cell library with seemingly no error.
For reference, it would require a group of 10 people to work for over a year to port a new technology cell library. The same can be accomplished with the help of a few powerful Nvidia GPUs in a matter of days, explained Dally. Of course, human intervention is still required in all these areas, as futuristic as it sounds. However, the AI-enabled Nvidia GPUs help the company massively save time and make a better-designed chip. In addition to GPUs, Nvidia might soon dip its toes in the CPU manufacturing business as well, but that’s a story for another time.
Source: Nvidia

#Nvidia #Faster #GPUs


Synthetic: Vik News

Đỗ Thủy

I'm Do Thuy, passionate about creativity, blogging every day is what I'm doing. It's really what I love. Follow me for useful knowledge about society, community and learning.

Trả lời

Email của bạn sẽ không được hiển thị công khai. Các trường bắt buộc được đánh dấu *

Back to top button