Все публикации

Serving Very Large0:08:39

Serving Very Large Models on K8s with Leader Worker Set

Access GCP Resources0:07:39

Access GCP Resources Securely with Workload Identity Federation for GKE

Serve Llama 3.10:04:09

Serve Llama 3.1 405B on Kubernetes on Multi Host GPUs

GKE Time Sharing0:04:39

GKE Time Sharing for GPUs

GPU Sharing in0:05:38

GPU Sharing in GKE with NVIDIA MPS

Improve Infrastructure Autoscaling0:08:52

Improve Infrastructure Autoscaling with Custom Compute Classes in GKE

GPU Sharing on0:05:16

GPU Sharing on GKE with Multi Instance GPU

Different ways of0:05:43

Different ways of Running RayJob on Kubernetes

Simplify Kuberay with0:03:48

Simplify Kuberay with Ray Operator on GKE

GKE Multi Tenancy0:14:34

GKE Multi Tenancy with Teams

Fleet Level Feature0:11:30

Fleet Level Feature Management with Feature Manager

Build Internal Developer0:12:07

Build Internal Developer Platforms on GKE using GKE Enterprise

Tips for Securing0:08:37

Tips for Securing your Ray Cluster on GKE

Effective GPU Sharing0:05:35

Effective GPU Sharing Strategies in GKE

Serving Gemma on0:04:32

Serving Gemma on GKE on TPU using Jetstream

Improve Resource Obtainability0:08:04

Improve Resource Obtainability (GPUs, TPUs) with Dynamic Workload Scheduler on GCP

Reducing data pre-processing0:08:59

Reducing data pre-processing time by 95% using Ray

Serving Gemma on0:05:24

Serving Gemma on GKE using Nvidia TRT LLM and Triton Server

Serving Gemma on0:05:43

Serving Gemma on GKE using Text Generation Inference (TGI)

Serving Gemma on0:04:56

Serving Gemma on GKE using vLLM

Improve LLM accuracy0:19:50

Improve LLM accuracy and performance with Retrieval Augmented Generation

Monitoring ML Training0:06:53

Monitoring ML Training Platform using Kueue Metrics and Cloud Monitoring

AI/ML on GKE:0:08:18

AI/ML on GKE: 2023 A Year in Review

Architecture of a0:16:41

Architecture of a ML Platform with Resource Sharing on Kubernetes