Solely per week after Nvidia’s new AI-focused Volta GPU structure was introduced, Google goals to steal a few of its thunder with its new, second-generation, Tensor Processing Unit (TPU) that it calls a Cloud TPU. Whereas its first technology chip was solely appropriate for inferencing, and due to this fact didn’t pose a lot of a risk to Nvidia’s dominance in machine studying, the brand new model is equally at house with each the coaching and working of AI techniques.
A brand new efficiency chief amongst machine studying chips
At 180 teraflops, Google’s Cloud TPU packs extra punch, not less than by that one measure, than the Volta-powered Tesla V100 at 120 teraflops (trillion floating level operations per second). Nonetheless, till each chips can be found, it received’t be potential to get a way of an actual world comparability. Very similar to Nvidia has built servers out of multiple V100s, Google has additionally constructed TPU Pods that mix a number of TPUs to realize 11.5 petaflops (11,500 teraflops) of efficiency.
For Google, this efficiency is already paying off. As one instance, a Google mannequin that required a whole day to coach on a cluster of 32 high-end GPUs (in all probability Pascal), could be skilled in a day on one-eighth of a TPU Pod (a full pod is 64 TPUs, so which means on eight TPUs). After all, commonplace GPUs can be utilized for all kinds of different issues, whereas the Google TPUs are restricted to the coaching and working of fashions written utilizing Google’s instruments.
You’ll be capable to lease Google Cloud TPUs to your TensorFlow functions
Google is making its Cloud TPUs obtainable as a part of its Google Compute providing, and says that they are going to be priced just like GPUs. That isn’t sufficient data to say how they are going to examine in price to renting time on an Nvidia V100, however I’d count on it to be very aggressive. One disadvantage, although, is that the Google TPUs at present solely assist TensorFlow and Google’s instruments. As highly effective as they’re, many builders won’t wish to get locked into Google’s machine studying framework.
Nvidia isn’t the one firm that needs to be anxious
Whereas Google is making its Cloud TPU obtainable as a part of its Google Compute cloud, it hasn’t stated something about making it obtainable outdoors Google’s personal server farms. So it isn’t competing with on-premise GPUs, and positively received’t be obtainable on aggressive clouds from Microsoft and Amazon. The truth is, it’s more likely to deepen their partnerships with Nvidia.
The opposite firm that ought to in all probability be anxious is Intel. It has been woefully behind in GPUs, which implies it hasn’t made a lot of a dent within the quickly rising marketplace for GPGPU (Common Function computing on GPUs), of which machine learning is a big half. This is only one extra method that chip that might have gone to Intel, received’t.
Massive image, extra machine studying functions shall be shifting to the cloud. In some circumstances — should you can tolerate being pre-empted — it’s already inexpensive to lease GPU clusters within the cloud than it’s to energy them regionally. That equation is simply going to get extra lopsided with chips just like the Volta and the brand new Google TPU being added to cloud servers. Google is aware of that key to rising its share of that market is having extra vanguard software program working on its chips, so it’s making 1,000 Cloud TPUs obtainable without cost to researchers prepared to share the outcomes of their work.