在 Amazon EKS 上运行高性能生成式 AI应用.pdf

《在 Amazon EKS 上运行高性能生成式 AI应用.pdf》由会员分享，可在线阅读，更多相关《在 Amazon EKS 上运行高性能生成式 AI应用.pdf（34页珍藏版）》请在三个皮匠报告上搜索。

1、 2025,Amazon Web Services,Inc.?*?务?/?2025,Amazon Web Services,Inc.?*?务?/?Amazon EKS?AI?2025,Amazon Web Services,Inc.?*?务?/?AI?2025,Amazon Web Services,Inc.?*?务?/?AI?2023?2024?2025?AI?Prompt?Agent?AI?2025,Amazon Web Services,Inc.?*?务?/?2023,Amazon Web Services,Inc.?研?研?AI?2025,Amazon Web Services,Inc

2、.?*?务?/?EKS?AI 2025,Amazon Web Services,Inc.?*?务?/?AI(Generative AI)?ML?1.?/?EC2?NeuronUltraClustersEFANitroGPUInferentiaSageMakerTrainium2.?|Agent|?Amazon Bedrock3.?Amazon Q Business Amazon Q DeveloperAmazon Q in Amazon QAmazon Q in Connect?(FM)?2025,Amazon Web Services,Inc.?*?务?/?EKS?AI模?ML?Kubern

3、etes?AI?的?的?2025,Amazon Web Services,Inc.?*?务?/?GPUAmazon InferentiaAmazon Trainium?EKS?/?GPU?Karpenter?ML?ML?2025,Amazon Web Services,Inc.?*?务?/?Amazon EKS?100?70?AI?20?EKS?2,000?CPU?GPU?Inf2?GPU?EKS?务 150?3?GPU?Hugging FaceHugging Face Hub-AI?2025,Amazon Web Services,Inc.?*?务?/?EKS AI/ML?2025,Amaz

4、on Web Services,Inc.?*?务?/?EKS?AI?Amazon EKS?/?Amazon S3?/?Amazon ECRAmazon AuroraAmazon NeuronAmazon EBSAmazon EFSAmazon FSx?NVIDIA GPU?RESTAPI 2025,Amazon Web Services,Inc.?*?务?/?Kubernetes?Amazon S3 Mountpoint CSI?S3?Kubernetes?Amazon S3?Mountpoint CSI?S3?Kubernetes Pod?Amazon S3?Mountpoint EKS?M

5、ountpoint CSI?EKS?Mountpoint?S3?2025,Amazon Web Services,Inc.?*?务?/?Kubernetes?ML?ML?GPU?Kubernetes?务 ML?够?研?GPU?EC2?2025,Amazon Web Services,Inc.?*?务?/?EKS?kubelet?CNI?CPU?Kubernetes?Kubernetes 务?Pod 务?EKS 务?GPU?GPU?2025,Amazon Web Services,Inc.?*?务?/?Amazon CloudWatch Container Insights?Neuron?NVI

6、DIA?Elastic Fabric Adapter 务?EKS 务?研?ML?研?2025,Amazon Web Services,Inc.?*?务?/?EKS?2025,Amazon Web Services,Inc.?*?务?/?EKS?NVIDIA GPUInferentiaTrainiumNeuronRuntimeNVIDIA?Neuron?+?EFA?EFA?NVIDIA?Neuron?EFA?EFA?vLLMRayCUDARuntime 2025,Amazon Web Services,Inc.?*?务?/?x86arm64?Savings PlansSpot?GravitonA