This week in Google Cloud — “Welcome to us-east4, GPUs for ML, inside TPUs, and your data via client libs”
Following a previous post quantifying the performance of a Tensor Processing Unit (TPU), this past week offered an in-depth look at Google’s first…