Google I/O 2024 once again showed that Google is focussing heavily on artificial intelligence. In addition to numerous software announcements, the focus was also on hardware innovations and infrastructure improvements that have the potential to significantly change the AI landscape.

Trillium: the sixth generation of TPU

One of the most significant announcements was the presentation of Trillium, the sixth generation of Google’s Tensor Processing Units (TPUs). These chips, specially developed for AI workloads, are designed to deliver an impressive increase in performance. With up to 4.7 times higher peak performance and double the memory bandwidth compared to the previous generation TPU v5e, Trillium promises to train and execute sophisticated AI models even more efficiently. Trillium is also more energy efficient by over 67%, which makes an important contribution to sustainability.

Google I/O 2024 Präsentation von Trillium

Axion: Google’s first ARM-based chip

Another milestone is the Axion processor, Google’s first self-developed chip based on the Neoverse V2 design of the ARM architecture. This chip was specially developed for the Google Cloud data centres and is intended to provide a significant increase in performance. According to Google, Axion offers up to 30% more performance than the fastest general-purpose ARM-based instances in the cloud and is up to 50% faster and 60% more energy-efficient than comparable x86-based instances.

NVIDIA Blackwell: Expansion of the partnership for more AI power

Google Cloud is growing its partnership with NVIDIA to integrate the new Blackwell GPU architecture into its cloud platform. Blackwell promises a significant increase in performance for AI and high-performance computing applications and will be used in various areas in the coming years. Integration into Google Cloud will enable customers to utilise Blackwell’s performance for their own AI projects. Initial benchmarks indicate an up to 30-fold increase in inferencing performance and up to 25-fold improvement in energy efficiency for large language models (LLMs) compared to conventional GPUs. In addition, the 5th generation of NVLink enables GPU-to-GPU communication at up to 1.8 TB/sec, which is twice as fast as the 4th generation used by Hopper.

AI hypercomputers and sustainable infrastructure

Google also emphasised its progress in the area of AI infrastructure at its I/O conference. The AI Hypercomputer, a supercomputing architecture for AI, combines CPUs, GPUs and TPUs to achieve twice the efficiency of conventional hardware and chips. This impressive increase in performance is made possible by a combination of different factors:

  • Hardware: high-performance CPUs, GPUs and TPUs at the heart of the AI hypercomputer.
  • Open source software: The use of open source frameworks such as SAX, TensorFlow, PyTorch and Kubernetes Engine accelerates and simplifies the development and deployment of AI applications.
  • Flexible utilisation: Dynamic workload schedulers, CUDA and on-demand spot instances enable efficient use of resources and optimal adaptation to the respective requirements.
  • Liquid Cooling: With a capacity of one gigawatt, the innovative cooling technology contributes to high energy efficiency by effectively cooling the high-performance hardware. However, it should be noted that this technology involves high water consumption, an aspect that should not be ignored in the discussion about sustainability (keyword: Green IT).

The infrastructure also includes Google Cloud’s fibre optic network, which is almost ten times longer than that of the next largest cloud provider at 2 million miles. This enables fast and reliable connections for AI applications and other cloud services.

Outlook

The hardware innovations and infrastructure improvements presented at Google I/O 2024 show that Google continues to invest heavily in the development of AI-specific hardware and infrastructure. With Trillium, Axion, the expanded partnership with NVIDIA for Blackwell and the powerful AI hypercomputer, Google is positioning itself as a leading provider of AI infrastructure and enabling companies and developers to take full advantage of the benefits of artificial intelligence. It will be interesting to see what further developments in this area will follow in the near future.

Get in touch

Your DataCentre hardware maintenance solution provider

Thanks to decades of experience, we know what is important when it comes to supporting and maintaining your data centre hardware. Benefit not only from our experience, but also from our excellent prices. Get a quote and compare for yourself.
Get in touch

What is Hardwarewartung 24 thinking?

Although OpenAI snatched some media attention from Google with its “Frontrunning” event on Monday, for hardcore AI experts, Google’s position is still unbroken. When it comes to multimodal LLM capabilities and agent development, Google is right up there with its competitor OpenAI. Both have presented a personal assistant that can communicate with people in real time (with a delay of less than 250 ms) via text, audio or camera and appears to understand its environment.

However, users are still very sceptical that these are just “showcases” that no one has yet been able to test in real life. Only the launch of the OpenAI app with ChatGPT-4 omni or Google’s Astra will show which assistant can deliver what it promises.

Subject: Hardware

Developments in the hardware sector are clearly noticeable and show the arms race for the future of the value chain through artificial intelligence. The performance data of the 6th TPU generation is particularly noteworthy. Although Google is expanding its partnership with NVIDIA and plans to offer Blackwell chips to its customers by 2025, the new TPU generation signals that Google is no longer willing to pay NVIDIA’s inflated prices due to scarcity. Instead, they are developing their own, more powerful chips. Google is thus joining the ranks of hyperscalers to at least partially force NVIDIA’s hands.

The use of the new ARM-based Axion chip also comes as no surprise. After both Amazon and Microsoft included ARM-based chips in their portfolio due to their better energy efficiency, Google is now also following this trend.

Focus: IT infrastructure

The arms race to expand the AI infrastructure requires not only massive expenditure on computing power and storage, but also the expansion of the network infrastructure. With over 2 million miles of fibre optic cable, Google also has a competitive advantage here, which will benefit them particularly in the use of multimodal models.

Looking at the entire infrastructure that Google has built and plans to expand further, the question arises as to how Google intends to organise this very highly resource-intensive project sustainably. The presentation talked about liquid cooling, which is very energy-efficient but neglects the issue of water consumption. To cool one gigawatt of power with water, entire rivers are needed, which are then to be offset with water compensation projects or “water stewardship”. However, these compensation projects have a similar aftertaste to CO2 compensation projects.

Summing up

Google showed at the I/O presentation that they are in no way inferior to the competition in terms of technology. What they have indirectly admitted, however, is that they had a lot of catching up to do when it came to integrating the technology into their products and are now catching up. They obviously don’t just want to benefit from the AI hype through the Google Cloud, but also integrate AI into all their products in order to utilise a wide range of value creation through AI. What we find problematic about this development is that the consumption of resources through AI is not being addressed. Google, as one of the pioneers of Green IT, would have had a good position here, but has obviously made a conscious decision not to address the issue.

Even if sustainability and Green IT are currently lagging behind due to the rapid development of AI, we believe that AI can also solve any problem that it has created itself.

Request a Quote

Your Maintenance Specialist for All Major Hardware OEMs

Thanks to decades of experience, we know what is important when it comes to hardware support and maintenance for your data centre. Benefit not only from our experience, but also from our excellent prices. Get a quote and compare for yourself.
Request a Quote