\n<\/p>\n
Google is making its own Arm-based CPU to support its AI work in data centers and introducing a more powerful version of its Tensor Processing Units (TPU) AI chips. Google\u2019s new Arm-based CPU, dubbed Axion<\/a>, will be used to support Google\u2019s AI workloads before it rolls out to business customers of Google Cloud \u201clater this year.\u201d<\/p>\n<\/div>\n The Axion chips are already powering YouTube ads, the Google Earth Engine, and other Google services. \u201cWe\u2019re making it easy for customers to bring their existing workloads to Arm,\u201d says Mark Lohmeyer, Google Cloud\u2019s vice president and general manager of compute and machine learning infrastructure, in a statement to Reuters<\/em><\/a>. \u201cAxion is built on open foundations but customers using Arm anywhere can easily adopt Axion without re-architecting or re-writing their apps.\u201d<\/p>\n<\/div>\n Google says customers will be able to use its Axion CPU in cloud services like Google Compute Engine, Google Kubernetes Engine, Dataproc, Dataflow, Cloud Batch, and more. Reuters <\/em>reports that the Axion Arm-based CPU will also offer 30 percent better performance than \u201cgeneral-purpose Arm chips\u201d and 50 percent more than Intel\u2019s existing processors.<\/p>\n<\/div>\n Google is also updating its TPU AI chips that are used as alternatives to Nvidia\u2019s GPUs for AI acceleration tasks.\u00a0\u201cTPU v5p is a next-generation accelerator that is purpose-built to train some of the largest and most demanding generative AI models,\u201d says Lohmeyer<\/a>. A single TPU v5p pod contains 8,960 chips, which is more than double the amount of chips found on the TPU v4 pod.<\/p>\n<\/div>\n Google\u2019s announcement of an Arm-based CPU comes months after Microsoft revealed its own custom silicon chips<\/a> designed for its cloud infrastructure. Microsoft has built its own custom AI chip to train large language models and a custom Arm-based CPU for cloud and AI workloads. Amazon has also offered Arm-based servers for years through its own custom CPU, with the latest workloads able to use Graviton3 servers on AWS.<\/p>\n<\/div>\n Google won\u2019t be selling these chips to customers, instead making them available for cloud services that businesses can rent and use. \u201cBecoming a great hardware company is very different from becoming a great cloud company or a great organizer of the world\u2019s information,\u201d says Amin Vahdat, the executive in charge of Google\u2019s in-house chip operations, in a statement to The Wall Street Journal<\/em><\/a>.<\/p>\n<\/div>\n