Buzz
At the recent GTC event, NVIDIA unveiled Project Earth-2, the world's most powerful AI supercomputer, a 3D simulation of Earth with flexible and detailed modeling capabilities to help humans assess climate change more accurately and predict future natural disasters.
Earth is heating up at an alarming rate. The past 7 years can be considered the hottest in human history. Since the period from 1850 - 1900, human-induced greenhouse gas emissions have caused the planet's average temperature to rise by about 1.1 degrees Celsius per year. What we are experiencing is truly alarming, with a slew of extreme weather events such as prolonged droughts, heavy rains, devastating hurricanes, and catastrophic floods. Natural disasters, which have always been harsh, are now becoming even more brutal and relentless.
However, addressing climate change is neither simple nor achievable overnight. Just a few years, or even decades, are not enough for us to feel positive changes. However, for Earth-saving campaigns to proceed in the most proper manner, we need to foresee the future, to sense what will happen with the highest possible accuracy.
It may sound like science fiction, but that's what NVIDIA is pursuing. At the GTC event held in mid-November last year, they unveiled Project Earth-2 (E-2), the world's most powerful AI supercomputer, the 'twin brother' of our Earth, on the Omniverse platform, a 3D simulation of the blue planet capable of providing more accurate predictions about climate change.
According to CEO Jensen Huang, E-2 will harness a combination of 3 cutting-edge technologies: high-speed GPUs, deep learning, and AI supercomputers, along with a massive Earth database. All of these will support the creation of climate prototypes with ultra-high resolution and the highest accuracy.
In reality, there are already numerous models like E-2 existing, capable of determining factors such as air pressure, wind intensity, and temperature to generate appropriate equations, providing an objective view of climate types in specific regions. These regions will be represented in a 3D grid format. The smaller the area, the more accurate the calculations, and the more realistic the simulation before it becomes unusable.
In other words, weather models need to solve more equations to achieve higher resolutions. However, receiving more equations makes the model slower, less efficient, and gradually less useful. That's the issue most current climate models are facing: a lack of both detail and accuracy.
And the solution NVIDIA proposes is a bigger, better, faster supercomputer. On the company's blog, Mr. Huang wrote: 'We need higher resolutions to simulate changes in the global water cycle. We need meter-scale resolutions to simulate the process of sunlight being reflected by clouds, bouncing back into space. Computational scientists estimate that such resolutions require computing power millions to billions of times stronger than what we have currently'.
Returning to E-2, Earth's digital twin is created to promote actions that both alleviate climate change and minimize its negative impacts on nature and humanity. Extreme weather phenomena like storms, wildfires, heatwaves, or floods are becoming increasingly unpredictable, with more devastating effects.
This situation could be somewhat improved if we could predict such disasters more accurately in the future. Mr. Huang hopes NVIDIA's model can forecast extreme weather changes in various regions worldwide decades in advance. Then, humans would have time to devise timely coping strategies, reasonable evacuation policies, or suitable infrastructure projects to adapt well to such climate conditions.
E-2 is also utilized to find rational and cost-effective solutions, simulating various plans to determine which one yields the highest efficiency with the lowest cost. This is also considered one of NVIDIA's largest projects in many years. Mr. Huang stated: 'All the technologies we have innovated so far are necessary to realize E-2. I am really looking forward to the new features, the more important tasks that this model can undertake in the future'.
Not only limited to climate change
The combination of various technologies has helped NVIDIA create the most advanced and efficient Earth simulation, while also addressing many issues related to the speed of supercomputers, especially in research projects with massive databases.
Similar to E-2, NVIDIA has focused on developing 3 core technologies: High-performance computing, AI, and the scale of data centers. This not only helps them simulate Earth, but also creates many digital twins of cities, factories worldwide. This large-scale simulation technology is still very new and full of potential.
Furthermore, the research teams at NVIDIA, Cal Tech, and the startup Entos have successfully combined machine learning with physics to create the OrbNet program. As a result, Entos can accelerate the simulation process of exploring their new drugs by up to 1000 times, completing the workload in over 3 months in just about 3 hours.
According to VentureBeat, NVIDIA
