Building Supabase Storage. DevX, performance, security - choose three
summary
Supabase builds an object storage service that stores file blobs in managed S3-compatible backends and keeps object metadata inside the user’s Postgres database, exposing access via PostgREST with JWTs and Postgres role‑level security. Defaults change: 50 MB cap on uploads and a one-hour cache by default to improve performance. Schema uses bucket name as primary key (UUIDs dropped) and public access requires signed URLs. API targets S3-compatible six ops so new backends are easy to add.
AI-written summary · about
chapters (12)
- 0:00Introduction to Supabase
- 1:05Supabase storage overview
- 2:05Design constraints and choices
- 3:35Selecting the storage backend
- 5:34Architecture and optimizations
- 7:30Database schema design
- 8:41Security and public access
- 9:32Metadata management
- 11:05Authorization with Postgres
- 14:32Internal API implementation
- 15:36Integration and extensibility
- 16:38Client API design and wrap-up
related talks
20mDesigning Supabase Edge Functions
17mA new trend for databases: Platforms
30mAmazon S3 foundations: Best practices for Amazon S3
13mSystems @Scale Summer Live Q&A Session #3
38mDelivering Enterprise-Grade Object Storage: Bloomberg’s IAM Journey in a Multi-Tenant Arch...
31mModern Security with Microservices and the Cloud
01h 03m3 innovations that redefine data protection for Amazon S3
47mMy entire stack is in my Postgres DB and a few power tools for my backend