What you need to know
- Elon Musk’s xAI team reportedly set up a powerful supercomputer featuring 100,000 H200 Blackwell GPUs in 19 days.
- NVIDIA CEO Jensen Huang claims the same process would ordinarily take up to 4 years to execute, prompting him to call Elon Musk “superhuman.”
- Huang also indicated that NVIDIA played a significant role in the process, offering their networking expertise to xAI as NVIDIA gear is complicated compared to traditional data center servers.
As a tech enthusiast who’s witnessed the rapid evolution of AI and its applications, I find myself utterly amazed by Elon Musk’s latest feat. The man has once again proven that he operates on a different plane when it comes to engineering and project execution. Setting up a supercomputer featuring 100,000 H200 Blackwell GPUs in just 19 days is something I would’ve thought possible only in a sci-fi novel, not real life!
Previously, Elon Musk, Tesla’s proprietor, disclosed that xAI has initiated the process of educating Grok AI utilizing the globe’s most potent AI system, which comprises 100,000 liquid-cooled NVIDIA H100 graphics processing units (GPUs) linked through a unified RDMA network. By year’s end, this powerful cluster is expected to make Grok AI’s chatbot the most powerful AI system in every aspect.
Now, it appears that the billionaire’s ambitious plan has successfully materialized. In a recent gathering with Tesla owners in Silicon Valley, NVIDIA CEO Jensen Huang commended Elon Musk’s hard work. This was in response to setting up a supercluster consisting of 100,000 H200 Blackwell GPUs within just 19 days, which is being used to drive advancements in xAI.
Huang explains that setting up xAI’s supercomputer was an intricate and challenging process, best handled by experienced specialists. He also notes that it’s typically expected for such a project, from inception to completion, to span as long as 4 years. However, Elon Musk managed to accomplish this in just 19 days. According to the CEO, this timeframe is divided into three years of preparation and a year dedicated to procuring equipment, installation, and ensuring the entire project runs smoothly.
According to NVIDIA CEO Jensen Huang:
From what I understand, it seems that only one individual possesses such unique skills – Elon. His grasp of engineering, construction, system management, and resource mobilization is exceptional, almost unimaginable.
The project involved building the necessary infrastructure for the high-performance supercomputer, complete with liquid cooling and power systems to accommodate advanced components such as 200,000 GPUs. Notably, NVIDIA’s CEO, Jensen Huang, mentioned that his team worked closely with xAI on various aspects of the project, including engineering, networking, infrastructure, and software development. He acknowledged that setting up the network for this sophisticated equipment can be more intricate than in conventional data center servers.
More Prime Day deals and anti-Prime Day deals
Read More
- SQR PREDICTION. SQR cryptocurrency
- DOGS PREDICTION. DOGS cryptocurrency
- LDO PREDICTION. LDO cryptocurrency
- DOP PREDICTION. DOP cryptocurrency
- HI PREDICTION. HI cryptocurrency
- METIS PREDICTION. METIS cryptocurrency
- EVA PREDICTION. EVA cryptocurrency
- UXLINK PREDICTION. UXLINK cryptocurrency
- KNINE PREDICTION. KNINE cryptocurrency
- CLOUD PREDICTION. CLOUD cryptocurrency
2024-10-17 21:10