Question Sourcing NVIDIA H100 Chips: Will the company be able to source NVIDIA H100 chips?

Answer: I have gone though some research material and I have found out some insight into this topic:

A) Partnership with NVIDIA:

1. NVIDIA lists three types of partnerships on their website: ELITE, Preferred, and partner.

2. There are a total of 73 cloud partners for NVIDIA. Among these, 19 are elite partners, and 61 are preferred partners.

3. In India, there are only two cloud partners, both of which are preferred partners: E2E Networks Limited and Yotta Data Services Private Limited.

   Advantages of being a preferred NVIDIA partner:

• Sales and Distribution Support: Preferred partners gain access to enhanced sales and distribution resources, enabling them to reach more customers effectively.

• Training Opportunities: NVIDIA provides specialized training programs, allowing partners to build expertise in GPU-accelerated computing and stay up-to-date with the latest technologies.

• Marketing Collaboration: Preferred partners can collaborate on marketing initiatives, co-branding efforts, and joint campaigns to promote their solutions.

• Service and Support: These partners receive priority support, ensuring timely assistance for technical queries and customer needs.

• Recognition and Rewards: NVIDIA acknowledges their commitment by recognizing Preferred partners and rewarding their achievements.

B) Commitment done by Jensen Huang –

• India is getting special attention. In September’2023, Huang met with Prime Minister Narendra Modi and said he would prioritise any orders from data centre operators in country. “You have the data, you have the talent,” Huang said at the time. “This is going to be one of the largest AI markets in the world.”

• “In pursuit of this goal, Yotta is acquiring 16,000 of Nvidia’s H100 Tensor GPUs. After all, Nvidia practically owns 90% of the GPU market. This acquisition is noteworthy considering the global shortage of Nvidia H100 GPUs, which has caused delays in numerous major projects worldwide. Despite the challenges posed by the global GPU shortage, the breakthrough in securing these GPUs for Yotta followed Prime Minister Narendra Modi’s meeting with Nvidia CEO Jensen Huang.”

• “In September, when Jensen Huang, founder & CEO, Nvidia came to India, he had met Prime Minister Narendra Modi. At that time, Prime Minister had expressed his desire of bridging this gap, helping India come at the forefront of AI, and even sought Huang’s help for the same. The Nvidia CEO offered PM Modi to help in his best capacity possible by offering technology and support to the local providers,” says Sunil Gupta, Co-founder & CEO of Yotta. He adds, “As somebody has to invest in, take a step forward to build this infrastructure in India, Yotta took the first step and placed the order.”

  Read More at:
  NVIDIA-Yotta deal marks India’s biggest AI bet yet – CNBC TV18

C) Geopolitical relations –

• Nvidia and India have a shared interest in betting on, and speeding up, the country’s AI ascendancy. Chipmakers cannot sell top-end microprocessors to China, which accounts for a fifth of Nvidia’s sales, amid fears the chips could be used to develop autonomous weaponry or wage cyberwarfare.

• “India is the only market remaining so it isn’t surprising that Nvidia wants to put multiple eggs in that basket,” said Neil Shah, vice president of research at Counterpoint Technology Market Research.

  Read more at:

D) Sovereign AI –

• Sovereign AI refers to a nation’s capabilities to produce artificial intelligence using its own infrastructure, data, workforce, and business networks. Essentially, it means that a country aims to rely on itself for AI development and usage, without external dependencies. This concept transcends technology; it’s about a country’s identity and autonomy. By building its own AI capabilities, a nation can strategically protect and advance its interests in an increasingly AI-driven world.

• NVIDIA has openly advocated for democratizing AI and has talked about the importance of sovereign AI at various occasions and conferences. This will help all the nations in building its own AI capabilities starting from data centers infrastructure, cloud computation to using its own data to create its own models and AI.

E) Small Quantity required compared to total Nvidia GPU production –

• In Year 2023 the company sold around 550,000 of its latest H100 compute GPUs worldwide in 2023, with most going to American tech firms. Looking ahead to 2024, Nvidia plans to at least triple the output of its compute GPUs, including the H100, GH100, and other processors for artificial intelligence (AI) and high-performance computing (HPC) applications. The projected shipments for 2024 range between 1.5 million and 2 million units.

• Despite this growth, the E2E network requirement remains minuscule compared to the total number of units produced.

@akk_07 : Sorry for the delayed response. I dont visit this forum often

• Waiting for history to repeat – This screener compares the current bull run (2013 onwards) to the 2003-2008 bull run. For instance, if a stock went up 100x between 2003 to 2006, but has moved up only 2x from 2013, then the ratio (100/2 = 50) is used to rank the stock. So that way you can find stocks that did well in that bull run but are yet to perform now. I have been using this screener since 2017 to build my own portfolio.
• Long Term trend reversal – This screener is based on long term averages of the 200-day RSI. For uptrend, it calculates the 200-day RSI over the entire pricing history of the stock, and if the current value if greater than 90% of all such values then it marks the stock as trending upwards. Similarly for downtrend, if the current value is in the bottom 20% of all 200-day RSI values for that stock then mark it as downtrend.

All the screeners use only pricing history, and price data is updated every weekend.

For catching trends early, I added Stock and sector trends. The first chart is interactive – you can click on a sector the pie chart and it will show matching stocks in “Recent trending stocks”. You could also subscribe to my Telegram channel.

But there is no perfect method – if you catch a trend too early you end up stuck with an underperforming stock and if you enter late then you miss the initial jump in prices. I have found better luck with identifying trending sectors and finding other underpriced stocks in that sector.

Would you plz share the rational for these and the exit triggers?

Thanks

I had bought at 350 and sold at 1000 as it was looking expensive to me at 1000​:joy: