跨区域部署 AI 模型确保弹性和合规性.pdf

1、Cross-Region AI Model Deployment for Resiliency and ComplianceTony Farias&Greg WoodJune 10th,2025Forward-looking StatementThis presentation has been prepared for informational purposes only.The information set forth herein does not purport to be complete or contain all relevant information.Statement

2、s contained herein are made as of the date of this presentation unless stated otherwise.This presentation and the accompanying oral commentary may contain forward-looking statements.In some cases,forward-looking statements can be identified by terms such as“may”,“will”,“should”,“expects”,“plans”,“an

3、ticipates”,“could”,“intends”,“projects”,“believes”,“estimates”,“predicts”,or“continue”,or the negative of these words or other similar terms or expressions that concern Databricks expectations,strategy,plans,or intentions.Forward-looking statements are based on information available at the time thos

4、e statements are made and are inherently subject to risks and uncertainties that could cause actual results to differ materially from those expressed in or suggested by the forward-looking statements.Forward-looking statements should not be read as a guarantee of future performance or outcomes.Excep

5、t as required by law,Databricks does not undertake any obligation to publicly update or revise any forward-looking statement,whether as a result of new information,future developments or otherwise.2AgendaResiliency&redundancy for AI ModelsGeodistributed AI models on DatabricksExample&Demo:Multi-Regi

6、on Serving on DatabricksEnd-to-end WalkthroughConsiderations for prod-hardened systemsKey Takeaways3What well cover todayResiliency&Redundancyfor AI ModelsWhy worry about model distribution?GPU Availability:Popular cloud regions often hit bottlenecks on VM types,especially“latest&greatest”SKUsData R