Valkey

Valkey is Watasu’s managed in-memory data service. It speaks the Redis protocol, so any Redis client library works against it without changes.

Use Valkey for fast, ephemeral state. Use PostgreSQL when the data must be your long-term system of record.

When to reach for Valkey

• caching computed values or query results
• session storage
• rate limiting and throttling
• ephemeral coordination (locks, leader election with TTL)
• queue acceleration

Create

watasu addons:create valkey --app my-app

Pass a plan to pick a tier:

watasu addons:create valkey:hobby-1 --app my-app

Plans range from hobby-0 through premium-3, differing on memory size, persistence, and whether backups are included. See Add-on Plans for the full catalog.

Connect from your app

After attachment, your app gets:

Variable	Purpose
REDIS_URL	Redis-protocol connection URL

The variable name keeps the Redis prefix on purpose so existing libraries and frameworks “just work.”

Open an interactive session

watasu valkey:cli --app my-app

The legacy redis:cli alias still works for muscle memory.

Backups

Valkey backups are available on persistent plans:

• entry plans: manual backups only
• higher plans: scheduled backups + manual

Capture

watasu addons:backups:capture my-cache

List

watasu addons:backups my-cache

Restore

Restores create a replacement add-on, then you promote it. Restore from a managed backup or from a local .rdb file:

watasu addons:restore my-cache ./backup.rdb --name my-cache-restored
watasu addons:promote my-cache-restored

For the full restore workflow, see Backups and Restores.

Tips

• Treat Valkey contents as expendable unless you’ve explicitly chosen a persistent plan with backups.
• Set TTLs on everything you can — unbounded keyspaces are how Valkey instances run out of memory at the worst possible moment.
• For pub/sub or fan-out patterns at scale, Redpanda is a better fit.