Google took a major step towards competing with Nvidia and mastering the AI models in cloud by introducing its Ironwood artificial intelligence chip, its most powerful TPU.
Google unveiled the seventh generation of its dedicated TPU processors, Ironwood at the Cloud Next event. Designed to deduce artificial intelligence models, the chip is twice as much as better performance and less energy consumption than its predecessor, Trillium. Google’s goal is to develop this chip, directly compete with Nvidia and provide high -scale processing power to Google Cloud users.
The Ironwood chip provides a maximum processing power of 4614 Traflas with 192 GB and 7.4 inserted Trabite bandwidth. This chip can be used in clusters with a configuration of up to 9216. One of its advanced features is the proprietary sparscore cores that make it possible to perform complex tasks such as scattered data analysis and content suggestion systems.
In the midst of the fierce competition of technology giants to reduce Nvidia’s dependence, Google with Ironwood has taken an important step in mastering artificial intelligence processing infrastructure. The chip is expected to be added to the Google Claud AI Hypercomputer and will play a key role in the future of the company’s AI infrastructure.
Latest Passing over countries : Spain | Dominica | United Arab Emirates
RCO NEWS
