Posts by tag

TPU

2 posts
View Post
  • 5 min

Google Cloud Unveils World’s Largest Publicly Available ML Hub With Cloud TPU V4, 90% Carbon-Free Energy

At Google, the state-of-the-art capabilities you see in our products such as Search and YouTube are made possible by Tensor Processing Units (TPUs), our custom machine learning (ML) accelerators. We offer these accelerators to Google Cloud customers as Cloud TPUs. Customer demand for ML capacity, performance, and scale continues to increase at an unprecedented rate. To support the next generation of fundamental advances in artificial intelligence (AI), today we announced Google Cloud’s machine learning cluster with Cloud TPU v4 Pods in Preview — one of the fastest, most efficient, and most sustainable ML infrastructure hubs in the world. Powered by…
View Post
Share
View Post
  • 11 min

PyTorch/XLA: Performance Debugging On TPU-VM Part 1

In this three part series we explore the performance debugging ecosystem of PyTorch/XLA on Google Cloud TPU VM. TPU VM earlier this year (2021). The TPU VM architecture allows the ML practitioners to work directly on the host where TPU hardware is attached. With the TPU profiler launched earlier this year, debugging your PyTorch training on TPU VM is simpler than ever before. While the process to analyze the performance has changed, the fundamentals of PyTorch/XLA that you have acquired with the network attached TPU architecture (aka TPU Node architecture), still apply. In this (first) part we will briefly lay…
View Post
Share