Skip to main content

Nvidia is powering a mega Tesla supercomputer powered by 10,000 H100 GPUs

Web Hosting & Remote IT Support

Tesla has revealed its investment into a massive compute cluster comprising 10,000 Nvidia H100 GPUs specifically designed to power AI workloads.

The system, which went online this week, is designed to process the mountains of data its fleet of vehicles collect with a view to accelerating the development of fully self-driving vehicles, according to its leader of AI infrastructure, Tim Zaman.

Tesla has been striving for years to reach the point at which its vehicles can be considered entirely autonomous and has invested more than a billion dollars into adopting the infrastructure to make this possible.

Tesla supercomputer

In July 2023, CEO Elon Musk revealed the firm would invest $1 billion into building out its Dojo supercomputer over the next year. Dojo, which is based on Tesla’s own tech, began with the D1 chip, fitted with 354 custom CPU cores. Each training tile module comprises 25 D1 chips, with the base Dojo V1 configuration including 53,100 D1 cores in total.

The firm also built a compute cluster fitted with 5,760 Nvidia A100 GPUs in June 2012. But the firm’s latest investment in 10,000 of the company’s H100 GPUs dwarfs the power of this supercomputer. 

This AI cluster, worth more than $300 million, will offer a peak performance of 340 FP64 PFLOPS for technical computing and 39.58 INT8 ExaFLOPS for AI applications, according to Tom’s Hardware

The power at Tesla’s disposal is actually more than that offered by the Lenoardo supercomputer, the publication pointed out, making it one of the most powerful computers on the planet.

Nvidia’s chips are the components that power many of the world’s leading generative AI platforms. These GPUs, which are fitted into servers, have several other use cases from medical imaging to generating weather models.

Tesla is hoping to use the power of these GPUs to more efficiently and effectively churn through the vast quantities of data it has to build a model that can successfully rival a human.

While many businesses would usually lean on infrastructure hosted by the likes of Google or Microsoft, Tesla’s supercomputing infrastructure is all on-prem, meanig the firm will also have to maintain all of it.

More from TechRadar Pro



via Hosting & Support

Comments

Popular posts from this blog

This new malware campaign can hijack your Gmail or Outlook email account

Web Hosting & Remote IT Support Cybersecurity researchers from Cisco Talos have spotted a new hacking campaign they claim is targeting victims’ sensitive data, login credentials, and email inboxes. Horabot is described as a botnet that has been active for almost two and a half years now (first spotted in November 2020). During that time, it’s mostly been tasked with distributing a banking trojan and spam malware .  Its operators seem to be located in Brazil, while its victims are Spanish-speaking users located mostly in Mexico, Uruguay, Venezuela Brazil, Panama, Argentina, and Guatemala. Horabot botnet The victims are found in different industries, from investment firms to wholesale distribution, from construction to engineering, and accounting. The attack starts with an email message carrying a malicious HTML attachment. Ultimately, the victim is urged to download a .RAR archive, which holds the banking trojan.  The malware is capable of doing plenty of things: stealing l

Want to store 1PB of data in the cloud? This startup can do it for you for as little as $10,000 a month — Qumulo says it can scale to Exabytes off premise and wants to eradicate tapes once and for all

Web Hosting & Remote IT Support Qumulo has launched Azure Native Qumulo Cold (ANQ Cold), which it claims is the first truly cloud-native, fully managed SaaS solution for storing and retrieving infrequently accessed “cold” file data. Fully POSIX-compliant and positioned as an on-premises alternative to tape storage, ANQ Cold can be used as a standalone file service, a backup target for any file store, including on-premises legacy scale-out NAS, and it can be integrated into a hybrid storage infrastructure, enabling access to remote data as if it were local. It can also scale to an exabyte-level file system in a single namespace. “ANQ Cold is an industry game changer for economically storing and retrieving cold file data,” said Ryan Farris, VP of Product at Qumulo. “To put this in perspective with a common use case, hospital IT administrators in charge of PACS archival data can use ANQ Cold for the long-term retention of DICOM images at a fraction of their current on-premises leg

No light without dark : making the most of ‘shadow IT’

Web Hosting & Remote IT Support In the last few decades, technology has created a modern digital workforce that is technically skilled and adept at finding innovative solutions that would help them succeed at work. However, with 95% of employees struggling with digital friction in the workplace - including a lack of access to the right tools - ambitious employees who are hungry for results have often needed to explore fixes outside the scope of existing systems provided by their employers. On top of that, the popularity of cloud-based apps has resulted in business processes often ending up fragmented across various systems, requiring workers to devote time to manual maintenance. This has accelerated the spread of (the unnecessarily ominous sounding) ‘shadow IT’, or applications that savvy workers use without official authorization to help them bypass limitations and get work done. In a perfect world, a balance can be struck between giving these technically skilled workers freed