Revolutionizing Business: Oracle’s Zettascale AI Leap

Key points

• Oracle Cloud Infrastructure (OCI) has announced the largest AI supercomputer in the cloud, OCI Zettascale10, which will serve as the backbone for the ambitious $500 billion Stargate project.
• The supercomputer will deliver unprecedented performance with up to 10X more zettaFLOPS of peak performance, making it ideal for complex computations like those calculated by advanced AI and machine learning (ML) systems.
• Oracle is now taking orders for OCI Zettascale10, which will be available in the second half of 2026, and is expected to benefit industry-specific AI applications with accelerated performance, enterprise scalability, and operational efficiency.

As a reporter, I have been following the latest developments in the world of AI, and I am excited to share with you the latest news from Oracle Cloud Infrastructure (OCI). This week, the company announced what it calls the largest AI supercomputer in the cloud, OCI Zettascale10. This multi-gigawatt architecture links hundreds of thousands of Nvidia GPUs to deliver unprecedented performance. The supercomputer will serve as the backbone for the ambitious, yet somewhat embattled, $500 billion Stargate project.

So, how does this supercomputer work? Oracle’s new supercomputer stitches together hundreds of thousands of Nvidia GPUs across multiple data centers, essentially forming multi-gigawatt clusters. This allows the architecture to deliver up to 10X more zettaFLOPS of peak performance, making it ideal for complex computations like those calculated by advanced AI and machine learning (ML) systems. To put that in perspective, a zettaFLOP (a 1 followed by 21 zeroes) can perform one sextillion FLOPS (floating point operations per second), allowing systems to perform intensely complex computations.

Oracle has also introduced new capabilities in Oracle Acceleron, its OCI networking stack, that helps customers run workloads more quickly and cost-effectively. These include dedicated network fabrics, converged NICs, and host-level zero-trust packet routing that Oracle says can double network and storage throughput while cutting latency and cost. The new architecture has a "wide, shallow, resilient" fabric, according to Oracle, and takes advantage of switching capabilities built into modern GPU network interface cards (NICs).

But is this supercomputer really necessary? While it seems like an astronomical amount of compute, there are customers for it, particularly envelope-pushing companies like OpenAI. These companies are pushing the boundaries of what is possible with AI, and they need the kind of power that OCI Zettascale10 can provide. Oracle is now taking orders for OCI Zettascale10, which will be available in the second half of 2026, and is expected to benefit industry-specific AI applications with accelerated performance, enterprise scalability, and operational efficiency.

As Alvin Nguyen, a senior analyst at Forrester, pointed out, most AI models have been trained on text, essentially at this point comprising "all of human-written history." Now, though, systems are ingesting large and compute-heavy files including images, audio, and video. "Inferencing is expected to grow even bigger than the training steps," he said. This means that companies will need more powerful computing systems like OCI Zettascale10 to handle these complex computations.

However, not every company can afford their own AI mega-factories. Nguyen advised companies to get creative and look for other ways to improve performance and speed, such as updating software stacks or analyzing their supply chain and supply chain management capabilities. They can also take advantage of mega-factories owned and run by others, such as Oracle’s OCI Zettascale10. As Nguyen pointed out, there is a growing "AI divide," but companies don’t have to be at the top to be successful. They just need to be able to execute when the opportunity arises.

In the world of AI, Microsoft and its companies, such as Azure, are also playing a major role. Azure is a cloud computing platform that provides a wide range of services, including AI and machine learning. Microsoft is also investing heavily in AI research and development, and its Azure platform is being used by many companies to deploy their AI models. As the demand for AI continues to grow, Microsoft and its companies are well-positioned to provide the necessary tools and services to support this growth.

Oracle’s announcement of OCI Zettascale10 is a significant development in the world of AI, and it will be interesting to see how Microsoft and its companies respond to this challenge. One thing is certain, however: the future of AI is going to be shaped by the ability of companies to provide powerful computing systems like OCI Zettascale10. As Yaz Palanichamy, senior advisory analyst at Info-Tech Research Group, pointed out, Oracle’s new supercomputer is designed to integrate large-scale generative AI use cases, including training and running large language models. This is an area where Microsoft and its companies are also investing heavily, and it will be interesting to see how these two companies compete in the future. With the growing demand for AI and the increasing need for powerful computing systems, the future of AI is looking bright, and Oracle’s OCI Zettascale10 is just the beginning.

Read the rest: Source Link

Don’t forget to check our list of Cheap Windows VPS Hosting providers, How to get Windows Server 2022, Try Windows 11 Pro for Workstations & browse Windows Azure content.

Remember to like our facebook and follow us on twitter @WindowsMode.