From Idea to Implementation: How Microsoft Achieved its ChatGPT…

From Idea to Implementation: How Microsoft Achieved its ChatGPT…

Facebook
Twitter
LinkedIn

  • microsoft corp MSFT pledged to build a massive, state-of-the-art supercomputer for OpenAI in 2019 when it invested $1 billion.
  • Microsoft had to find ways to string tens of thousands together The Nvidia Corp NVDA A100 graphics chips, the workhorse for training AI models, and changing the positioning of servers on racks, Bloomberg reports.
  • OpenAI has long needed access to robust cloud computing services as it attempts to train an ever-increasing number of AI programs, called models.
  • Scott Guthrie, Microsoft executive vice president, said it cost Microsoft over several hundred million dollars.
  • “We have built a system architecture that can work and be reliable at very large scale. That’s what led to ChatGPT becoming possible,” said Nidhi Chappell, Microsoft’s general manager.
  • “This is a model that came out of it. There will be many, many others.”
  • Microsoft uses the same resources it developed for OpenAI to train and run its own large AI models, including the new Bing search bot.
  • Microsoft is already working on the next generation of the AI ​​supercomputer, part of an expanded contract with OpenAI in which Microsoft added $10 billion.
  • Training a massive AI model requires a large pool of connected GPUs in one place, like the AI ​​supercomputer Microsoft has assembled.
  • For the AI ​​supercomputer, Microsoft had to develop software to get the best results from the GPUs and network equipment.
  • Microsoft also uses graphics chips for inference that are geographically distributed across the company’s more than 60 data center regions.
  • The company is adding the latest Nvidia graphics chip for AI workloads, the H100, and the latest version of Nvidia’s Infiniband networking technology to share data even faster.
  • Price promotion: MSFT shares traded 2.61% higher to $255.09 on the latest check Monday.

More to explorer