33 documentos etiquetados con "Google"

CPU, GPU, and TPU roles for AI training and inference

GCP AI and ML Services

GCP AI Hypercomputer for large-scale AI workloads

GCP Smart Analytics Services

Enable and manage Google Cloud APIs

Business continuity fundamentals and the cloud design choices that support recovery

Cloud TPU optimization principles and practical workload patterns

Cloud TPU architecture, scaling model, and core platform concepts

GCP Compute Services

Migrating databases to GCP

High-level design tradeoffs for cloud architecture.

Separating what a system must do from how well it must do it

First things you need to know

Google Cloud platforms for provisioning and running GPU clusters

CUDA, XLA, and framework portability patterns for GPU-accelerated AI workloads

ML Productivity Goodput and practical GPU optimization patterns for AI workloads

Google Cloud GPU machine families for AI training and inference

Initial setup for GCP

Common patterns for integrating systems in the cloud

Connecting business goals to measurable targets and architecture choices

How to design systems that are observable from the start

Organization policies in Google Cloud

Resource hierarchy in Google Cloud

GCP Storage Services

GCP streaming analytics with Pub/Sub and Dataflow

Cloud TPU consumption options, quota planning, and DWS scheduling modes

Cloud TPU deployment options and workload scaling patterns

Overview of the Cloud TPU family

TPU runtime version choices and practical TPU and GPU interoperability patterns

Cloud TPU architecture, scaling model, and core platform concepts

How to read Google Cloud scenarios by identifying the real business driver and product position

Google Cloud Well-Architected Framework for design and PCA decisions

How to choose the right environment for your workload