A library consortium needs common computer-vision checks on uploade…

Question

A library consortium needs common computer-vision checks on uploaded photos and wants strong performance with minimal GPU fleet operations. What should the architect prefer?

Accepted Answer

Prefer Amazon Rekognition for common vision tasks unless custom models require DIY GPUs — If the vision job is a common one, don’t stand up a GPU ranch—use Rekognition and save the ops. DIY GPUs wait for custom models that managed APIs can’t cover. Deep Archive yearly jobs and Config rules aren’t real-time vision engines.

Answer

Always build a self-managed GPU EC2 fleet first, even for standard vision APIs

Answer

Store images only in Glacier Deep Archive and run vision offline annually

Answer

Use AWS Config rules as the image classification inference engine

A library consortium needs common computer-vision checks on uploaded photos and wants strong performance with minimal GPU fleet operations. What should the architect prefer?

Short Explanation

Full Explanation