Google has unveiled a new version of its data center artificial intelligence (AI) chips and announced an Arm-based central processor, marking significant advancements in the company’s hardware offerings.
On Tuesday, Google introduced the latest iteration of its tensor processing units (TPUs), specialized chips designed to accelerate machine learning tasks. These TPUs are among the few viable alternatives to advanced AI chips made by Nvidia, offering enhanced performance for AI applications.
In addition to the AI chips, Google announced an Arm-based central processing unit (CPU) for its data centers. The move to Arm architecture signifies a shift toward more energy-efficient and customizable processors, potentially impacting the dynamics of data center hardware.
While Google’s TPUs are powerful, developers can currently access them only through Google’s Cloud Platform, and they are not available for direct purchase. This strategy positions Google to strengthen its cloud services by offering cutting-edge AI computing capabilities to its users.
Google’s latest hardware announcements highlight the company’s commitment to advancing AI technology and cloud computing infrastructure. These developments could have significant implications for businesses and researchers relying on high-performance computing resources.
Reference(s):
cgtn.com
