Repository files navigation

This repository is a collection of accelerated platform best practices,reference architectures, example use cases, reference implementations, andvarious other assets on Google Cloud.

An accelerated platform utilizes specialized hardware components, oraccelerators, such asGPUs (Graphics Processing Units) andTPUs (Tensor Processing Units), to significantlyincrease the speed of computationally intensive tasks. These tasks may includedata analysis, machine learning, artificial intelligence, and graphicsrendering. By offloading demanding workloads from traditional CPUs to dedicatedhardware accelerators, which are capable of much faster parallel calculations,the platform optimizes high-performance computing.

TheCloud Workstations (CWS) Platform is acore, best practices, fully managed workstation environments built to meet theneeds of security-sensitive enterprises. It enhances the security of workstationenvironments while accelerating onboarding and productivity.

• Cloud Workstations reference architecture
  • Cloud Workstations reference implementation

TheGKE Base Platform is an implementationof a foundational platform built on GKE that incorporates best practices andprovides a core environment optimized for running accelerated workloads. Itoffers a streamlined and efficient solution to leverage the benefits of GKE asthe primary runtime.

• ComfyUI reference implementation
• Federated learning
• Inference reference architecture
  • Inference reference implementation
    • Online inference with GPUs
      • Online inference using Diffusers with GPUs on Google Kubernetes Engine (GKE)
      • Online inference using vLLM with GPUs on Google Kubernetes Engine (GKE)
    • Online inference with TPUs
      • Online inference using MaxDiffusion with TPUs on Google Kubernetes Engine (GKE)
      • Online inference using vLLM with TPUs on Google Kubernetes Engine (GKE)
• Training reference architecture
  • Model fine tuning
    • Data processing
    • Data preparation
    • Fine tuning
    • Model evaluation

• LLM Inference Optimization: Achieving faster Pod Startup with Google Cloud Storage
• Optimizing GKE Workloads with Custom Compute Classes

ThePlayground AI/ML Platform on GKEis a quick-start implementation of the platform that can be used to familiarizeyourself with the GKE architecture and to get an understanding of variousconcepts covered in the use cases.

• GKE AI/ML Platform for enabling AI/ML Ops

• Scalable and Distributed LLM Inference on GKE with vLLM
• Retrieval Augmented Generation (RAG) pipeline

For more information about contributing to this repository, seeCONTRIBUTING.