Databases

Upstash

Serverless Redis and Kafka with global replication and REST API access.

Use Cases

Cache API responses in a serverless app with global read replicas for low-latency access

Set up a message queue between microservices using Kafka with serverless scaling

Free Tier

10MB Redis, 10K commands/day; 10K messages/day Kafka

How to Maximize the Free Tier

Upstash's free tier (10MB Redis, 10K commands/day) works for caching and session storage, not primary data storage. Keep your keys small and set TTLs aggressively — expired keys don't count toward storage. The REST API is the killer feature: use it from edge functions where Redis protocol isn't available. For the Kafka free tier (10K messages/day), use it for lightweight event queues, not high-throughput streaming.

Getting Started

Sign up → create Redis database → copy REST URL and token → use `fetch()` or SDK to cache data globally → add Kafka cluster for message queues if needed. No server to manage.

Pros

REST API: Access Redis via REST API — no Redis protocol dependency, works from edge functions and browsers
Global replication: Read replicas in multiple regions for low-latency data access from anywhere in the world
Pay-as-you-scale: Pricing scales with usage after the free tier — no upfront commitments or minimum spends

Cons

Redis size: 10MB Redis is tiny — sufficient for session caching but inadequate for any real data storage
Command quota: 10,000 commands/day on the free tier can be tight for active applications with frequent cache operations
No persistence: No data persistence guarantees on the free tier — data loss possible during infrastructure events

Alternatives

Redis Cloud7/10★★★½☆Confluent Cloud